A STAR Research April-September 2014

HIGHLIGHTING THE BEST OF A STAR
FEATURES
Sustainable ways to keep us ying p3
Improving drug delivery for breast
cancer treatment p7
Technology on the catwalk p14
HIGHLIGHTS
Cell Biology & Immunology p15
Chemistry & Materials p29
Genetics & Disease p47
Physical & Life Science Technologies p59
Engineering & Nanotechnology p73
APRIL 2014 SEPTEMBER 2014
www.astar-research.com
A*STAR Research

ISSN 2010-0531
www.astar-research.com a-star_research@a-star.edu.sg
A*STAR Research is a publication of the Agency for Science,
Technology and Research (A*STAR) Singapores lead govern-
ment agency for fostering world-class scientic research.
A*STAR Research is published twice a year, presenting
research highlights and feature articles. All articles are rst
published online on the A*STAR Research website and app,
and available free to all readers. Register online to receive the
table of contents by email as each biweekly online issue
is published.
2014 A*STAR. This publication may be reproduced in its
original form for personal use only. Modication
or commercial use without prior permission
from the copyright holder is prohibited.
Editorial
Agency for Science, Technology and Research
1 Fusionopolis Way, #20-10 Connexis North Tower
Singapore 138632, Singapore
Editor-in-Chief David Lane
Editorial Board Huck Hui Ng
Colin Stewart
Andy Hor
Keith Carpenter
Chandra Verma
David Wu
Alex Gu Yuandong
Guan Cuntai
Boris Lukyanchuk
Managing Editors Kostas Repanas
Lia Paola Zambetti
Administrative Assistant Lay Hoon Tan
CELL BIOLOGY & IMMUNOLOGY
The secretions of stem cells 16
Helping cultured stem cells get back to nature 17
Focusing on the sweet spot 18
A roadmap for pure organ cells 19
The fat source makes the diference 20
Subverting the sneeze 21
Building a heart 22
Fishing for new therapeutics 23
Detailing the development of red blood cells 24
Putting the squeeze on the Singapore sneeze 25
Animal-free reprogramming improves safety 26
Skin cells control their own fate 27
Lab on breathing chip 28

CHEMISTRY & MATERIALS
Ensuring solid-state drives are up to scratch 30
Multilayer polymers spring into action 31
Probing polarization puzzles 32
The direct approach to microcavities 33
Microbeads are easily xed 34
A transistor for light 35
Finding the right mix 36
Making it big 37
Cubic cluster chills out 38
Minimizing loss by thinning and smoothing 39
Metalorganic micromushrooms repel all 40
Polycarbonates to tackle multidrug resistance 41
Dyes help harvest light 42
Powered by nanoholes 43
Modern routes to ancient remedies 44
Positive progress on antifouling 45
Mixing through oscillations 46
GENETICS & DISEASE
The power of proteins to battle the bulge 48
Flu vaccine enables rapid response 49
Stapling together cancer therapy 50
Antifungals go nano 51
The signicance of silence 52
Putting production in proportion 53
Finding consensus on target genes 54
The yin and yang of moonwalking cells 55
Structure informs function 56
A view of tumor vessels 57
The genes behind the guardians of the airways 58
PHYSICAL & LIFE SCIENCE
TECHNOLOGIES
Sweet sensing 60
Slimmed down for a better t 61
Grading cataracts automatically 62
Catalytic upgrade 63
Keeping in contact 64
Modied drug gives a green light for its own success 65
A click toward localized chemotherapy 66
Lighting up lymph nodes 67
Finding the sweet spot for cartilage formation 68
Small particles, big productivity 69
Sugarcoating a protein drug 70
Quick to recognize 71
Exercising restraint to stall tumor growth 72
ENGINEERING &
NANOTECHNOLOGY
Silicon helps light go through the right channels 74
Self-powered wireless light detectors 75
Enabling next-generation wireless networks 76
A modern twist on Youngs slits 77
More speed, less interference 78
The promise of purple for enhanced bioimaging 79
Clear predictions 80
Right on schedule 81
Recharging in private 82
Focusing on whats relevant 83
Cutting reection for the infrared and beyond 84
Getting the edge on tool wear 85
Getting the most out of tiny lasers 86
Hot spots for molecules 87
Microparticles get the whole picture 88
Protecting laptops on the move 89
Paving the way for electronic applications 90
Cracks emerge in the cloud 91
Modeling city growth from the ground up 92
ON THE COVER
Microfabrication
techniques developed at
A*STAR are being used to
manufacture safer neural
probe arrays [p61]
Sustainable ways to keep us ying 3
Harnessing randomness to improve lasers 5
Nourishing health in early life 6
Improving drug delivery for breast cancer treatment 7
Shoddy sorting disrupts memory-making signals 8
Setting sequencing free 9
Life on the big screen 11
Collaborating to boost the microuidics industry 12
Changing the face of Singapores infocommunications 13
Technology on the catwalk 14
RESEARCH HIGHLIGHTS
FEATURES & INNOVATIONS
Contents
HIGHLIGHTING THE BEST OF A*STAR
ASTAR_Collection_2012_09_Dividers.indd 1 28/09/2012 12:25PM
Download the A*STAR Research app
Clear. Concise. Convenient.
www.research.a-star.edu.sg/mobile
FEATURES &
INNOVATIONS
A*STAR RESEARCH APRIL 2014 SEPTEMBER 2014 www.astar-research.com 3
>>> 3
Research at A*STAR is helping to
ensure a sustainable future for the
aviation industry.
The global aviation industry continues to expand,
with over 3 billion people expected to fly com-
mercially in 2014, along with 38 million metric tons
of cargo. This activity will have a huge impact on the
environment and requires vast resources. In order
to make air travel a sustainable option for future
generations, the industry needs to perform continual
research into technologies that reduce both economic
and environmental costs.
The A*STAR Aerospace Programme was set
up to combine the efforts of A*STAR researchers
who are striving to provide a sustainable future for
aviation. Since 2007, scientists from several A*STAR
research institutes have worked together on over
50 multidisciplinary projects to pioneer manufac-
turing techniques, safety inspection devices and
analytical methods to improve flight management.
To highlight this success, their work was showcased
earlier this year at the Singapore Airshow 2014.
Finding faults faster
Aircraft will always experience wear and tear,
and A*STAR researchers are developing fast,
cost-effective methods to identify parts in need of
repair. Crucially, these methods aim to be non-
destructive, which means that fewer parts need to be
removed or replaced during inspection, saving both
time and money.
One major problem for the industry is the infiltra-
tion of water into the aircraft body, especially in
the lightweight honeycomb structures found in the
tail and wings. Dmitry Isakov from the A*STAR
Singapore Institute of Manufacturing Technology
(SIMTech) is leading efforts to help technicians spot
exactly where the water is.
Water always finds the easiest way to get in,
which is around discontinuities such as joints and
bolts, says Isakov. Once inside, the water expands
and contracts as it freezes and melts, damaging
structures, causing corrosion and increasing the
aircrafts weight.
To identify areas where water has entered, techni-
cians currently heat the aircraft surface and then use
thermal cameras to observe its cooling. Regions that
cool too quickly indicate the presence of water below
the surface. The method requires two engineers,
is slow and cannot distinguish between water and
excess sealant left behind after repairs.
Isakov has developed an alternative approach.
When a vacuum is created around a bolt head, water
can boil even at room temperature, he explains.
Boiling consumes heat, causing the material around
the water to cool down by as much as several degrees.
Water detection using my vacuum method requires
just one technician, is fast and highly sensitive, and
there is no ambiguity with the sealant, Isakov adds.
Building the future of ight
In addition to improving fault detection, researchers
at A*STAR are providing new methods for repairing
damaged aircraft parts or manufacturing new
designs. One promising technology under develop-
ment is laser-aided additive manufacturing (LAAM)
(see Laying the groundwork for a manufacturing
revolution). This method, an example of three-
dimensional printing technology, uses a high-energy
laser beam to deposit materials in precisely controlled
geometric structures or to fill in cracks.
Multidisciplinary research at A*STAR is producing new technologies to improve safety and
efciency in the aviation industry
Sustainable ways to keep
us flying

r
o
n
n
i
e
c
h
u
a
/
i
S
t
o
c
k
/
T
h
i
n
k
s
t
o
c
k
TECHNOLOGIES
FEATURES &
INNOVATIONS
www.astar-research.com A*STAR RESEARCH APRIL 2014 SEPTEMBER 2014 4
The bottom-up approach is less wasteful than
traditional methods that involve cutting components
out of larger chunks of material.
LAAM can repair or fabricate parts with excellent
mechanical properties and resistance to wear and
corrosion, says Guijun Bi, a leading researcher at
SIMTech. Bi is adapting LAAM techniques to build
and repair structures using so-called superalloys,
which maintain their strength even under the
extreme conditions of a working jet engine.
The precise control provided by LAAM is enabling
Bi and his co-workers to re-use components that
would previously have been impossible or very costly
to repair. The remanufacturing of components is
central to the aims of the A*STAR Aerospace Pro-
gramme and provides clear environmental benefits
through reducing resource consumption.
Keeping problems at bay
Other high-tech maintenance tools being developed
at SIMTech include a system that uses electromag-
netic waves to detect slight variations in the thickness
or composition of components that may be the result
of corrosion. Further monitoring can be provided
by using piezoelectric sensors that hear structural
failures in real time, with the added benefit of
turning the mechanical energy of the aircraft body
into electricity.
The lifetime of aircraft components can be greatly
extended by applying advanced water- and ice-repel-
lent superhydrophobic coatings developed
at SIMTech that protect parts from condensation,
corrosion and mold, as well as improving aerody-
namics to reduce fuel consumption. Researchers are
also pioneering the use of lasers, instead of expensive
corrosive chemicals, to strip damaged coatings from
turbines, allowing the parts to be re-used.
Improving the human factor
Aside from the development of inspection and
remanufacturing tools, diverse expertise in computer
simulation at A*STAR is providing novel ways to
improve the experience of airline customers and staff.
Anyone who has flown will have heard the cabin
crew asking passengers to switch off their electronic
devices during take-off and landing because stray
signals can interfere with flight systems. Now,
researchers at the A*STAR Institute for Infocomm
Research (I
2
R) have written simulations that identify
the best way to arrange electronics on the aircraft for
minimizing interference, enabling airlines to provide
customers with wireless services without compro-
mising safety. Another useful software developed
at the I
2
R, named Super De-haze, provides pilots
and air-traffic controllers with clearer imagery by
removing the effects of haze, fog and smoke.
Finally, the complex challenge of airport manage-
ment is being tackled by I
2
R researchers who have
developed a flight prediction algorithm that captures
the interactions of weather and flight congestion.
Their algorithm won first prize in the GE Flight
Quest 2013 challenge, estimating flight arrival times
that were 40 per cent more accurate than current
industry estimates. By applying similar models,
airlines could achieve more efficient operation of
gates and runways.
Flying ahead of the competition
The A*STAR Aerospace Programme is a prime
example of the benefits of multidisciplinary research.
The program is attracting interest from aviation
giants such as Airbus and Boeing, and is set to ensure
that Singapore maintains its competitive edge in the
global aerospace marketplace.
More importantly, by improving aircraft design,
maintenance, fuel economy and customer satisfac-
tion, A*STAR is leading the way toward a more
sustainable and environmentally friendly business
model for aviation. Only through such long-sighted
efforts will future generations continue to enjoy the
many benefits of air travel.
Vacuum of Vacuum on
W
a
t
e
r
N
o

w
a
t
e
r

2
0
1
4

A
*
S
T
A
R

S
i
n
g
a
p
o
r
e

I
n
s
t
i
t
u
t
e

o
f

M
a
n
u
f
a
c
t
u
r
i
n
g

T
e
c
h
n
o
l
o
g
y
Dmitry Isakov at A*STAR is investigating
the infiltration of water under the surface
of an aircraft by applying a vacuum to
a bolt head. Thermal imaging reveals a
cooler area around the bolt head when the
vacuum is switched on (bottom right) that
indicates the presence of water, which is
not seen in a dry bolt head (top right).
Laser-aided additive manufacturing (LAAM) (top) can be used to
create complex structures such as airfoils (bottom) used in
the development of aircraft wings from high-strength alloys.

2
0
1
4

A
*
S
T
A
R

S
i
n
g
a
p
o
r
e

I
n
s
t
i
t
u
t
e

o
f

M
a
n
u
f
a
c
t
u
r
i
n
g

T
e
c
h
n
o
l
o
g
y
FEATURES &
INNOVATIONS
Randomly arranged items usually have poor optical
properties. The rough or random surface of a
frosted-glass window, for example, obscures the view
of an object. The optical industry therefore expends
considerable effort reducing any surface irregularities in
optical devices to avoid the uncontrollable scattering of
light characteristic of random structures.
But now, a research group led by Ying Zhang from
the A*STAR Singapore Institute of Manufacturing
Technology (SIMTech) has made good use of
randomness by studying how random structures can
improve the performance of lasers. Together with a
team led by Qijie Wang at Nanyang Technological
University in Singapore, the group has demonstrated
the worlds first electrically pumped mid-infrared
random laser, which operates at a 10-micrometer
wavelength. The laser is as bright as conventional
diode lasers but produces less-speckled images.
Light waves from a conventional laser oscillate in
perfect synchronicity, across both time and space.
Perfect alignment of the light waves at different times
and different locations across the beam profile is known
as temporal and spatial coherence, respectively. When a
laser illuminates a surface, a speckled pattern is typically
visible, which indicates spatial coherence. The speckles
result from the laser beam reflecting from different
parts of the surface. Because the waves are in sync,
they create spatial interference effects in the eye of an
observer. This distortion is undesirable, particularly
in biomedical imaging applications conducted in the
infrared region of the spectrum.
Random lasers are the solution to this type of distor-
tion, says Zhang. Random lasers show the same high
temporal coherence as that of other lasers but have a
lower spatial coherence, he explains. High temporal
coherence gives the desirable brightness, but it is the
low spatial coherence that removes the speckles caused
by interferences.
To realize a random laser in the mid-infrared
spectrum, Zhang and co-workers used a semiconductor
quantum cascade laser into which they had drilled a
random pattern of nanoholes. At a sufficiently high
density, these holes prevent the formation of a regular
laser pattern within the semiconductor (see image).
Instead, the pattern of a random laser forms, with low
spatial coherence.
Employing a quantum cascade laser to realize the
random lasers allows for the polarization of the laser
light perpendicular to the laser surface. This propaga-
tion minimizes losses owing to the air-hole structure.
The research teams wafer-fabrication competencies
enabled them to drill holes deep enough into the laser
chip, with sufficiently smooth side walls to minimize
losses in the laser itself. By introducing these perfections
and overcoming a number of other practical hurdles,
Zhang and his colleagues succeeded in making
the lasers efficient enough to provide lasing during
electrical operation.
Nevertheless, notes Hou Kun Liang of SIMTech,
who invented the mid-infrared random laser, more
work is needed to bring random lasers to market.
We are working on a random laser that operates at
room temperature. And in the long-term, we plan to
extend random lasers from the infrared to even longer
wavelengths, where they can penetrate materials and
be used for inspection of various polymer packaging
quality-control of printed electronics, biomedical
imaging and other applications.
Developed at A*STAR, the rst electrically pumped random lasers for mid-infrared radiation
are set to enable applications in sensing and imaging
Harnessing randomness to
improve lasers

2
0
1
3

W
I
L
E
Y
-
V
C
H

V
e
r
l
a
g

G
m
b
H

&

C
o
.
K
G
a
A
,
W
e
i
n
h
e
i
m
Computed optical light fields in a
random laser overlaid on black circles,
which represent the nanoholes drilled
into a semiconductor quantum cascade
laser, to produce a laser pattern with
low spatial coherence.
TECHNOLOGIES
FEATURES &
INNOVATIONS
>>>
Nourishing health in early life
The A*STAR Genome Institute of Singapore is partnering with Nutricia Research in a clinical
and genomic study of the bacteria inhabiting the infant gut
As a newborn takes its first breath, microbes that line
the mothers birth canal are already making their way
into the infants gut. And with the babys first taste of
breast milk, many more will settle in. This process of
accommodating a friendly assortment of bacteria in
the digestive tract collectively referred to as the gut
microbiota is essential for healthy development and
helps the infant to digest foods, synthesize vitamins and
enzymes and fight pathogenic invasion.
Understanding the factors that affect the develop-
ment of microbial populations in babies is the aim of a
new partnership between the A*STAR Genome Insti-
tute of Singapore (GIS) and Nutricia Research, part of
the international food company Danone. The investiga-
tion will involve a clinical and genomic study of the gut
microbiome, which is the sum of the genomes of all the
microbes that reside in the gut. This partnership is
combining the genomic expertise of the GIS, who have
developed high-throughput approaches to monitoring
the microbiome, with the early life nutrition and clinical
expertise of Nutricia Research to investigate the health
benefits of bacterial supplements, says Martin Hibberd,
associate director at the GIS.
The researchers hope their findings could lead to
improved infant nutrition for better health in early, and
even adult, life.
Complex and resilient
A harmonious relationship with the bacteria that share
our digestive tracts is increasingly being recognized as
essential for good health.
Early-life perturbations of the developing gut micro-
biota can have an impact on the immune, metabolic and
neurological systems, and potentially lead to long-term,
adverse health outcomes, says Kaouther Ben Amor,
senior team leader for gut microbiology and physiology
at Nutricia Research. Studies have shown that a rift
in the relationship between microbes and the body
can contribute to diseases in humans such as allergy,
asthma, obesity, diabetes, inflammatory bowel
syndrome and even autism. But while previous focus
lay on identifying the patterns of microbial colonization
associated with disease states, the new endeavor should
help to define what a healthy bacterial population
looks like filling a gap in our current knowledge.
A healthy microbiome has not yet been characterized
in full, explains Hibberd. We do know, though, that
it is complex and in the study we will be seeking to
associate this with healthy outcomes. Researchers have
determined that a healthy bacterial community is not
only diverse but also ecologically stable, which means
that it can maintain its community structure in the
face of stress or rebound to the default state following a
disruption, adds Ben Amor.
Milk for health
Understanding the dynamics of the guts microbial
ecosystem and the factors that drive its development
can be used to help optimize health in later life. Of
particular importance is nutrition, in which the infants
mother plays a lead role. Human milk is normally the
first dietary exposure in infancy, and it is considered the
best nutrition for growth and healthy development of
the newborn, says Ben Amor.
Human milk contains two groups of compounds that
are considered beneficial for health prebiotics and good
bacteria that are also found in probiotic supplements.
Prebiotics are compounds that stimulate the growth
of specific good bacteria in the large intestine. These
include the nondigestible carbohydrates known as
oligosaccharides found in human milk. Probiotics are
live bacteria, such as species of the lactic-acid-producing
Lactobacillus and Bifidobacterium, which help maintain
microbial balance in the gut.
Researchers believe that a better understanding of the
contribution of prebiotics, probiotics and a combination
of the two synbiotics in fostering normal bacterial
populations could be used to improve the design of
nutritional products that supplement or complement
human milk.
Other factors also have an effect on the development
of an infant gut microbiome, says Hibberd. Antibiotics,
for example, may change the pattern of bacterial coloni-
zation. Without exposure to the bacterial populations in
the mothers birth canal, the gut microbiota of cesarean
babies more closely resembles the microorganisms found
on the surface of the mothers skin. Given the increasing
rates of cesarean deliveries worldwide, including in Asia,
more research is needed to investigate the implications
for health of the surgical delivery of babies.
Feeding bacteria
The GISNutricia Research partnership hopes to bring
new insights to these nutritional aspects of
Fluorescent in situ hybridization (FISH) is a
technique that uses fluorescent probes to
detect specific DNA sequences. Research-
ers will use FISH microscopy analysis
to determine the population dynamics
of Bifidobacterium bacteria present in
samples taken from study participants.

2
0
1
4

N
u
t
r
i
c
i
a

R
e
s
e
a
r
c
h
The Nutricia Research team led by Kaou-
ther Ben Amor (left), senior team leader
for gut microbiology and physiology.

2
0
1
4

N
u
t
r
i
c
i
a

R
e
s
e
a
r
c
h
FEATURES &
INNOVATIONS
Improving drug delivery for
breast cancer treatment
A nontoxic hydrogel developed by the A*STAR Institute of Bioengineering and
Nanotechnology and IBM Research ofers a new way forward for breast cancer therapeutics
Breast cancer is the most common form of invasive
cancer affecting women worldwide. Treatment
usually involves surgery followed by a course of either
chemotherapy, radiotherapy or hormone therapy
designed to reduce the risk of the cancers recurrence.
In recent years, however, targeted drugs have increas-
ingly become an effective option to fight the disease.
One such drug is Herceptin, also known as
trastuzumab, a monoclonal antibody that can slow
or even halt tumor growth in patients with human
epidermal growth factor receptor 2 (HER2)-positive
breast cancer a particularly fast-growing form of
the disease that affects one in four patients. Typi-
cally, the drug is administered at a clinic through
an intravenous drip, a process that can take up to
90 minutes. A major drawback of this method of
delivery is that without frequent follow-up doses,
Herceptin loses its effectiveness. Therefore, patients
are commonly required to visit the clinic on a
weekly basis.
Now, a team of researchers from the A*STAR
Institute of Bioengineering and Nanotechnology
(IBN) and IBM Research led by Yi Yan Yang and
James Hedrick has developed a more efficient way
of delivering Herceptin to breast cancer cells. The
approach promises to improve the effectiveness of
treatment for the HER2-positive form of the disease.
Drawing on IBNs expertise in the development
infant health. Taking inspiration from human milk,
it will look specifically at how the administration of
prebiotics, probiotics and synbiotics modulate the gut
microbiota in early life.
As part of the collaboration, Nutricia Research will
focus on conducting two clinical studies in Asia. The
studies will investigate how supplements of nutrition
can support establishment of the microbiome in early
life and help prevent disease later in life.
Samples taken from the gastrointestinal tracts of
study participants will be analyzed using the GISs state-
of-the-art genomic technologies and advanced analysis
techniques. This will allow researchers to characterize
the genomic content of the entire microbial community
and develop a molecular understanding of the benefits
associated with nutritional supplements.
Microbiome research is an exciting frontier and
it leverages on the GISs core capability to analyze
and study the complex community of microbes, says
Huck Hui Ng, executive director of the GIS. Only
recently, the GIS developed a method for identifying
the complement of species that constitute complex
microbial communities, which is more comprehensive
and cost-effective than existing techniques and can be
applied to large-scale studies.
A nutritional boost
Hibberd expects that the nutritional supplements
will contribute to an increase in microbial diversity
in the infant gut, thus generating health benefits for
the study population. With time, these findings could
translate into new nutritional products, specifically
designed to address the needs of target groups such
as babies born by cesarean delivery. Very little
science has been published on the health benefits of
different milk formulae in promoting healthy gut
microbiomes up to now, says Hibberd. We hope
that our study will benefit the field by generating this
new knowledge.
A team of researchers led by Martin
Hibberd (left), associate director at the
A*STAR Genome Institute of Singapore,
is partnering with Nutricia Research
in a clinical and genomic study of the
infant microbiome.

2
0
1
4

A
*
S
T
A
R

G
e
n
o
m
e

I
n
s
t
i
t
u
t
e

o
f

S
i
n
g
a
p
o
r
e
>>>
TECHNOLOGIES
FEATURES &
INNOVATIONS
>>>
of novel biomaterials, the researchers recognized
the advantages of using a hydrogel comprising
96 per cent water and a uniquely designed
polymer as a carrier for Herceptin in the body.
The hydrogel is nontoxic, biodegradable and
can be injected under the skin without causing an
inflammatory response. By varying the composition
and concentration of the polymer, the researchers
were able to fine-tune the hydrogels properties to
ensure a slow and sustained release of the anticancer
drug, thus increasing the efficiency of delivering the
drug to the target site. Correspondingly, use of the
hydrogel for drug delivery has the potential to reduce
the frequency at which patients need to be injected
with Herceptin from once a week to once every
four weeks.
Recently, the researchers conducted studies in
mice, which confirmed that their Herceptin-loaded
hydrogel delivers the drug efficiently, before it
degrades within 6 weeks. They found that 4 weeks
after injection, tumors in the mice had decreased in
size by as much as 77 per cent. The researchers say
that their next goal will be to conduct clinical trials
in humans, in conjunction with industrial partners.
Other teams at IBN are also exploring the
innovative use of hydrogels for controlled-release
drug delivery to tackle different types of cancer
(see Greater anticancer potency with less risk).
The institute has always cultivated a multi-pronged
approach toward cancer research, notes Jackie Y.
Ying, professor and executive director at IBN.
Our multidisciplinary research teams are working
with various industrial, clinical and academic part-
ners to develop new materials and tools to improve
cancer diagnosis and treatment, she adds. This
latest breakthrough with our long-term partner IBM
Research promises more efficient administration
of anti-cancer drugs and more effective treatment
of breast cancer, which we hope will benefit breast
cancer patients worldwide.
Shoddy sorting disrupts
memory-making signals
A*STAR scientists continue to explore the potential contributions of a sorting nexin protein
to learning impairments in Downs syndrome
Every year, roughly 1 in 1,000 children worldwide
are born with Downs syndrome. This developmental
disorder, associated with potentially severe intellectual
and learning disabilities among other character-
istics, is caused by the presence of a third copy of
chromosome 21, resulting in abnormal activity levels
for the more than 300 genes on this chromosome.
Scientists have had difficulty identifying the core genes
responsible for the disorder, but Wanjin Hongs team at
the A*STAR Institute of Molecular and Cell Biology in
Singapore have identified a gene with an important role
in brain signaling that is one possible culprit
1
.
Hong and colleagues focused their attention on
sorting nexin 27 (SNX27), one of a family of proteins
that coordinate movement of other proteins to different
compartments of the cell. Previously, the team showed
that SNX27 contains distinctive structural features
that suggested it may be active at neuronal synapses
2
.
In addition, mice lacking this protein displayed
severe developmental and neurological abnormalities,
including measurable cognitive defects.
SNX27-knockout mice demonstrated behavioral
characteristics that make them good candidates for
research on learning and memory, explains Li Shen Loo,
a research scientist in Hongs laboratory. Intriguingly,
this study also revealed that lower levels of SNX27 were
also apparent in brain tissue from patients with Downs
syndrome, and showed that SNX27 production is
reduced by one of the genes present on chromosome 21.
Loo and Hong decided to examine brain structure
and function in SNX27-deficient mice. They observed
considerable fluid accumulation within the mutant
mouse brain and underdevelopment of the dentate
gyrus a part of the brain responsible for learning
The sorting nexin SNX27 (green) is found
primarily at dendritic spines, as indicated
by overlapping fluorescence with postsyn-
aptic density protein-95 (PSD-95; red). R
e
p
r
o
d
u
c
e
d
,
w
i
t
h

p
e
r
m
i
s
s
i
o
n
,
f
r
o
m

R
e
f
.
1

L
.
S
.
L
o
o

e
t

a
l
.
When the anticancer drug Herceptin
is delivered to mice via a slow-release,
nontoxic hydrogel (left), it localizes to
tumors for a sustained period of at least
two weeks (right).

2
0
1
3

W
I
L
E
Y
-
V
C
H

V
e
r
l
a
g

G
m
b
H

&

C
o
.
K
G
a
A
,
W
e
i
n
h
e
i
m
FEATURES &
INNOVATIONS
and memory. These alterations made further analysis
tricky, according to Loo.
The knockout brains were too soft and watery to
characterize with traditional techniques, she says, but
I was able to investigate the mechanism of memory
impairment in these mice using live cell imaging
techniques. By examining different combinations of
fluorescently labeled proteins in neurons from wild-type
and SNX27-deficient mice, Loo and her colleagues
could directly observe the real-time behavior of indi-
vidual synapses in the presence and absence of SNX27.
Those experiments confirmed that SNX27 is pre-
dominantly found within dendritic spines (see image),
the part of the neuron that receives incoming signals at
the synapse. SNX27 specifically resides within struc-
tures called recycling endosomes, which help to shuttle
neurotransmitter receptors and other proteins between
the cellular interior and the cell surface.
When the researchers experimentally simulated
long-term potentiation, the neuronal activation process
associated with memory building, they observed
that SNX27-bearing endosomes moved from the
cell interior to the surface. These endosomes also
contained GluA1, a component of the receptor for the
neurotransmitter glutamate. In the absence of SNX27,
GluA1 is no longer efficiently transported to the
surface of the dendritic spines in fact, it becomes
susceptible to degradation. Since glutamateGluA1
signaling is critical for long-term potentiation, SNX27
could play a major role in enabling transmission of
memory-related signals.
These results could account for the learning deficits
observed in Downs syndrome, notes Loo, who sees
a potential therapeutic opportunity. Since Downs
syndrome patients produce less SNX27, re-introducing
the protein may restore memory function, she says.
Setting sequencing free
Researchers at the A*STAR Bioinformatics Institute develop a pioneering mobile application
for portable analysis of DNA sequences
A*STAR molecular cell biologist Samuel Gan was in
the midst of an exasperating work trip in Shanghai,
China. While he was away from his office, Gans
email inbox was filling up with DNA sequencing
files that needed his urgent attention.
Gans research group in Singapore is engaged in
therapeutic antibody production and had engineered
DNA molecules known as plasmids to transmit
protein-encoding information into cells. Before the
plasmids could be introduced into cells, however,
their DNA sequences had to be determined and
verified. But with no way to interpret these sequences
on his smartphone, Gan was unable to instruct
his team back home to begin the next stage of
producing antibodies.
I asked myself: Why cant I use my phone to do
something so simple? explains Gan, team leader of
the Antibody and Product Development Laboratory
at the A*STAR Bioinformatics Institute (BII). I
was frustrated.
Knowing there had to be a way to unlock the pro-
cessing power contained in his smartphone to decode
the sequencing files, Gan posed this challenge to a
research officer in his team, Phi Vu Nguyen. Within
three months, Nguyen had developed DNAApp
the first mobile application, or app, for viewing
and analyzing DNA sequencing files on an Android
mobile device. And since its April 2014 launch, close
to 700 scientists in over 11 countries have already
downloaded Android and iOS versions of the app.
Decoding DNA
DNA sequencing methods allow scientists to
determine the precise order of the four basic units,
or bases, of DNA adenine (A), guanine (G),
cytosine (C) and thymine (T) in a sample.
These methods have been used to sequence
DNA strands of varying lengths from individual
genes to the 3 billion base pairs of the human
genome. By analyzing this genetic information,
researchers can identify sequence mutations to
improve our understanding of the genetic causes
of a range of diseases, with the added possibility of
finding treatments.
Conventional sequencing technology employs
automated machines to record the information
contained in genetic material as raw data in the
Ab1 file format. Anyone who works on gene
sequences will be familiar with these files, explains
Gan. Until now, researchers have relied on conver-
sion programs designed for personal computers to
translate this data into more intelligible sequences of
As, Gs, Cs and Ts, which has prevented them from
completing sequence analysis while away from
DNAApp allows researchers to verify DNA
sequences and identify mutations on the go.

2
0
1
4

A
*
S
T
A
R

B
i
o
i
n
f
o
r
m
a
t
i
c
s

I
n
s
t
i
t
u
t
e
1. Loo, L. S., Tang, N., Al-Haddawi, M.,
Dawe, G. S. & Hong, W. A role for
sorting nexin 27 in AMPA receptor
trafficking. Nature Communications
5, 3176 (2014).
2. Wang, X., Zhao, Y., Zhang, X., Badie, H.,
Zhou, Y. et al. Loss of sorting nexin 27
contributes to excitatory synaptic
dysfunction by modulating
glutamate receptor recycling in
Downs syndrome. Nature Medicine
19, 473480 (2013).
>>>
TECHNOLOGIES
FEATURES &
INNOVATIONS
their desk. With DNAApp on the scene, scientists
can now undertake this task wherever they may be.
Mobile revolution
In Gans eyes, mobile applications are the next
frontier in helping scientists keep up with the
increasing demands of speedier science, especially in
the field of gene sequencing. Fifty to sixty years ago,
a researcher may have been able to publish a paper in
the journal Nature after cloning a particular restric-
tion enzyme, notes Gan. Today, even sequencing
a whole genome may not generate sufficient data for
publication in a high-impact journal.
The formerly labor-intensive and time-consuming
work of sequencing a gene can now be completed in a
day at an industrial scale and without human inter-
vention. Next-generation sequencing technologies
promise complete sequences of five human genomes
within a week, compared to the decade it took to
decipher the first human genome. Whole-genome
sequencing is now part of the norm, adds Gan. Our
standards have gone up and we need to move along
with those standards.
Until recently, however, mobile apps had not taken
advantage of the processing power and tools built
into smartphones in a way that really contributed to
research. Many existing apps are simply textbooks
pasted into an application, says Gan. DNAApp
could transform the way that scientists view their
phones as a tool for increasing productivity.
Screening sequences
DNAApp lets users work through DNA sequencing
tasks with a few simple gestures allowing
them to easily visualize sequences and assess the
quality of their samples. Researchers can copy,
cut and paste sections of genetic code, and even
search, locate and jump to sections of interest in
the sequence.
Another useful function of DNAApp is the
ability to convert a DNA sequence into its reverse
complement, replacing each base in a sequence with
its natural partner A with T, and G with C. This
feature is especially useful when initial sequencing
is performed on the DNA strand that pairs with the
strand of interest.
In addition, DNAApps translation feature
can interpret the sequence of bases as a chain of
corresponding amino acids, enabling a deeper
investigation of the effects of mutations, for example
to determine the association between small changes
in the DNA sequences that encode viral proteins and
drug resistance.
Users are already finding DNAApp to be an
essential tool. It has released me from being stuck in
front of a computer or a laptop, explains Gan. I can
do sequencing analysis on the go: on the bus, on the
train basically anywhere.
Feedback from those who have downloaded the
app has been extremely positive and in recognition
of the impact that the app is making, DNAApp has
been accepted for publication by leading journal
Bioinformatics, following the addition of extra
features and the development of a comprehensive user
guide. It has been rewarding to create something
that makes the life of a researcher easier and helps to
move their research forward, says Gan.
Advanced automation
Gan and Nguyen plan to develop further apps that
they hope will make the life of the experimental
scientist easier as well as establish the smartphone
as a convenient tool for the lab. Frank Eisenhaber,
executive director of the BII, has pledged his
continued support for these endeavors, stating that:
We will continue to develop creative ideas for useful
and efficient tools and techniques in computational
biology for applications in the life science field.
Smartphones, of course, take up only a slice of the
innovation potential in the broader field of computa-
tional biology, points out Gan. His dream is to have
robots perform experiments for him, so that he can
concentrate his efforts on the theoretical rather
than the operational aspects of his research. And
with advanced automation already built into a lot of
laboratory equipment, that reality may not be too far
off, he says.
Samuel Gan (right) and Phi Vu Nguyen
(left) from the A*STAR Bioinformatics
Institute have developed DNAApp
a convenient app for analyzing DNA
sequencing files on mobile devices.

2
0
1
4

A
*
S
T
A
R

B
i
o
i
n
f
o
r
m
a
t
i
c
s

I
n
s
t
i
t
u
t
e
FEATURES &
INNOVATIONS
Life on the big screen
A prize-winning microscopy image of a developing mouse sperm cell taken by A*STAR
scientists lights up Times Square in New York
Science and show business may sound like an unusual
combination, but advances in technology mean that
scientists can now capture dramatic images of their
research that easily match the glamor of Broadway.
A striking image captured by Graham Wright and
Henning Horn from the A*STAR Institute of Medical
Biology (IMB) in Singapore during their ground-
breaking investigations into fertility was the regional
winner in the microscopy category of the 2013 GE
Healthcare Life Sciences Cell Imaging Competition.
Fittingly, together with the other prize-winning images,
the image was recently displayed on a large, high-
resolution screen in New Yorks iconic Times Square.
The image is of a mouse sperm cell, also known
as a spermatocyte, highlighted with three fluorescent
labels that show DNA (blue), KASH5 protein (green)
and the SCP3 protein (red), which is required for the
pairing of chromosomes, explains Wright. This image
was a particularly striking example when we captured
it the orientation of the proteins we were studying
and the two sperm cells stained in blue on both sides
made it aesthetically pleasing.
The image was the result of a collaboration between
Wright, head of the IMB Microscopy Unit (IMU),
Horn, a senior research fellow, and the research teams
of Colin Stewart and Brian Burke, also of the IMB,
who discovered that the KASH5 protein is vital for
successful chromosomal movements during meiosis
the division of cells necessary for successful sexual
reproduction
1
. Sperm and eggs need accurate chromo-
some pairing if they are to mature correctly, so without
chromosomal activity guided by the KASH5 protein,
fertility is adversely affected.
The researchers collected the image on a GE
DeltaVision OMX microscope, which enables biological
samples to be imaged in superresolution in three
dimensions. Wright and Horn spent time perfecting
their sample preparation and honing the settings on
the microscope to acquire their high-resolution prize-
winning image. The increased resolution we were able
to achieve allowed us to visualize chromosome pairing
events in spermatocytes.
Paired chromosomes the paired red lines in
the image are not resolved by conventional light
microscopy techniques, explains Horn. The ability
to determine whether chromosomes are paired or
not was critical for understanding the function of
KASH5 in meiosis, so images such as this really can
change how we understand diseases and problems such
as infertility.
At the IMB, research is focused on understanding a
number of human diseases and medical conditions and
providing improved treatments. Researchers have direct
access to state-of-the-art equipment, including high-end
microscopes, through core technology platforms such
as the IMU.
By gaining further insight into the processes behind
meiosis using these advanced microscopy techniques,
Wright and Horn hope that their work will shed light
on a range of biological processes, including those
underpinning human fertility problems.
Microscopy is usually used to visualize and analyze
how proteins, cells and tissues are organized and how
they behave, states Wright. But with fluorescent dyes
and live-cell imaging techniques, we can find where
proteins are located and follow what they do over time.
This can give us excellent insights into the function of
a protein and what goes wrong with it in the diseases
we study.
The prize-winning image of a develop-
ing mouse sperm cell, captured by
A*STAR researchers Graham Wright
and Henning Horn. The locations of
DNA (blue), KASH5 protein (green) and
SCP3 protein (red) were revealed using
immunofluorescence staining.

2
0
1
4

A
*
S
T
A
R

I
n
s
t
i
t
u
t
e

o
f

M
e
d
i
c
a
l

B
i
o
l
o
g
y
1. Horn, H. F., Kim, D. I., Wright, G. D.,
Wong, E. S. M., Stewart, C. L. et al.
A mammalian KASH domain protein
coupling meiotic chromosomes to
the cytoskeleton. The Journal of Cell
Biology 202, 10231039 (2013).
TECHNOLOGIES
FEATURES &
INNOVATIONS
Collaborating to boost the
microfluidics industry
The A*STAR Singapore Institute of Manufacturing Technology establishes three new industry
collaborations to further innovation in microuidics
Diagnostics, pharmaceutical development and drug
delivery require manipulation of microscopic volumes
of liquid, and microfluidics is the science of producing
devices to do this. In a move that will speed up the
growth and development of the microfluidics industry,
the A*STAR Singapore Institute of Manufacturing
Technology (SIMTech) has recently sealed three
new collaborations with global companies, forming
partnerships that will lead to novel and cost-effective
microfluidic solutions.
The market for polymer-based microfluidic devices is
growing, with its value expected to reach US$2.7 billion
by 2018. SIMTech has already demonstrated its com-
mitment to the industry by establishing the SIMTech
Microfluidics Foundry (SMF) in 2011.
The SMF spearheads innovations in microfluidics
manufacturing technology and provides design,
prototyping and production services for microfluidics
development and applications, says SIMTechs director
of research programmes, Wang Zhiping. Through
various collaborations with industry, SMF nurtures
and grows the microfluidics industry by supporting the
business and research community.
The three new collaborations reflect the diversity
in SIMTechs microfluidics expertise. The first
collaboration is with specialist manufacturer InziGn,
in which SIMTech will transfer manufacturing
technology to the company, allowing mass produc-
tion of microfluidic devices.
InziGn has 30 years of experience in precision tool
making, particularly plastic injection mold fabrication,
says Steven Lau, InziGn director of product develop-
ment. This expertise in high-volume and high-quality
medical manufacturing is well positioned for mass
production of microfluidic devices. The agreement will
allow the SMF to continue supporting the development
of novel devices while InziGn manufactures those that
are ready to enter mass production for global markets.
The second collaboration sees the licensing of
SIMTech microfluidics technology to Singapore-
based biotechnology company Austrianova, which
encapsulates living cells in microdroplets of nonreactive
polymers. Austrianovas Cell-in-a-Box technology
allows the isolation, protection, storage and transporta-
tion of living cells and has wide-ranging applications,
from healthcare, to agriculture and environmental
protection. Following the agreement, Austrianova will
use SIMTech microfluidics-based droplet generation
technology to make the cell encapsulation process
more efficient.
SIMTechs droplet generator is made from a dispos-
able polymer chip, explains Austrianova CEO Brian
Salmons, meaning that sterilization is not required and
the downtime of the manufacturing line is reduced.
The final agreement initiates a research collabora-
tion between SIMTech and UK-based diagnostics
company QuantuMDx. We are leveraging SIMTechs
microfluidics expertise to develop portable point-of-
care assay cartridges for use as part of our handheld
molecular diagnostics (MDx) device, explains
QuantuMDx co-founder and chief scientific officer,
Jonathon OHalloran. The assay cartridges being
developed will provide microfluidics handling of
samples, on-chip sample preparation, polymerase chain
reaction capabilities and detection modules.
The MDx device will be able to diagnose disease
and detect drug resistance in less than 15 minutes, for
just a few dollars, by the patients side, says OHalloran.
Under the agreement, multiple cartridges will be
developed for the universal device to test for diseases
such as drug-resistant malaria, tuberculosis and sexually
transmitted diseases, as well as to perform tumor
profiling and companion diagnostics.
The device will be particularly useful in areas
lacking the healthcare infrastructure necessary for
traditional laboratories, though the devices low cost
and rapid turnaround times will also benefit developed
healthcare systems, he notes.
Microfluidic devices produced by A*STARs
SIMTech Microfluidics Foundry.

2
0
1
4

A
*
S
T
A
R

S
i
n
g
a
p
o
r
e

I
n
s
t
i
t
u
t
e

o
f

M
a
n
u
f
a
c
t
u
r
i
n
g

T
e
c
h
n
o
l
o
g
y
FEATURES &
INNOVATIONS
Singapore a knowledge-based economy known for
its business efficiency and global competitiveness is
fast becoming the information and communications
technology (ICT) capital of the world. In 2013, the
World Economic Forum ranked Singapore as the
most network-ready country in Asia, well ahead of its
neighbors Taiwan, South Korea and Hong Kong.
In view of rapid growth in Singapores infocommu-
nications sector, the A*STAR Institute for Infocomm
Research (I
2
R) has partnered with key players in
Singapores ICT economy to found REACH@I
2
R
(REsearch And Commercialization Hub), a cluster of
joint laboratories designed to nurture technological
innovation that meets the needs of multiple industries.
REACH@I
2
R brings together a diversity of scientific
capabilities to address the specific needs of local and
international companies, says Lee Shiang Long,
executive director of I
2
R. More than 130 research and
development personnel are already working in the joint
laboratories and their expertise is helping to accelerate
co-innovations with industry partners in finance,
energy, telecommunications, healthcare, transportation,
media and entertainment. REACH@I
2
R truly epito-
mizes our concept of marrying research with industry,
Lee affirms.
To ensure the greatest impact, Lee is leading
REACH@I
2
R with a system approach that harnesses
I
2
Rs diverse range of scientific capabilities and is
supported by several key initiatives, including the
creation of a vibrant ecosystem for ICT research that
incorporates multinational and globally competitive
companies, as well as small and medium enterprises.
REACH@I
2
R focuses on fostering ICT innovation
that brings value to Singapores economy, setting up
the necessary infrastructure and processes for tighter
integration with industry, as well as working with
local government to ensure sustained investment in
ICT in Singapore.
Already, the REACH@I
2
R cluster has produced
several original ICT solutions. In the field of healthcare,
these include a system for assessing visual acuity, per-
forming augmented-reality laser surgery and detecting
eye diseases, co-developed with Singapores National
Healthcare Group and Tan Tock Seng Hospital, as well
as technologies for detecting myopia, glaucoma and
age-related macular degeneration, in collaboration with
Japanese optical equipment manufacturer Topcon.
REACH@I
2
R is also making progress in the area of
communications, including the development of energy-
efficient networks for high-capacity communications,
in conjunction with Singapores ST Electronics. In
the area of human language and speech technologies,
REACH@I
2
R laboratories are creating software for
voice authentication, speech recognition, real-time
interpretation and automated dialog replacement.
One such example is a voice recognition software for
unlocking smartphones, developed with Chinese web
services company Baidu.
In the automotive industry, in conjunction with the
Chinese automobile and new-energy company BYD,
REACH@I
2
R is developing autonomous technologies
for driving vehicles remotely, activating brakes and
steering wheels. In the energy sector, smart grids for the
production and distribution of electricity with signifi-
cantly improved connectivity, security, stability and
intelligence are the focus of a REACH@I
2
R collabora-
tion with energy utility group Singapore Power.
Singapores positioning as an attractive gateway for
companies looking to grow internationally has made
A*STAR a partner of choice for the co-creation of
knowledge and innovation for the global market. We
have many companies that are proactively investing
in joint labs at REACH@I
2
R. They have faith in our
capabilities and decided to co-innovate with us on a sus-
tained basis to create a competitive edge for themselves,
notes Lee. Embracing research and development in this
way is set to positively impact the infocommunications
ecosystem in Singapore.
REACH@I
2
R, Singapores largest cluster of joint laboratories for infocommunications
research, is deepening links between A*STAR and industry
Changing the face of Singapores
infocommunications

2
0
1
4

A
*
S
T
A
R

I
n
s
t
i
t
u
t
e

f
o
r

I
n
f
o
c
o
m
m

R
e
s
e
a
r
c
h
Lim Chuan Poh, chairman of A*STAR,
officially opened the REACH@I
2
R joint
laboratories in March 2014.
Lee Shiang Long, executive director
of the A*STAR Institute for Infocomm
Research (I
2
R).

2
0
1
4

A
*
S
T
A
R

I
n
s
t
i
t
u
t
e

f
o
r

I
n
f
o
c
o
m
m

R
e
s
e
a
r
c
h
TECHNOLOGIES
FEATURES &
INNOVATIONS
Summer days bring thoughts of beach picnics, outdoor
barbecues and pool parties. Yet it only takes the buzz of
one tiny mosquito to dampen the fun.
But what if your outfit was equipped to release the
scent of lemongrass? A quick rub of your hands over
the fabric and the pests would be kept at bay by the
odor they so despise. This proposal was just one of
many wearable technologies caught walking down
the runway earlier this year at a showcase organized
by A*STARs technology transfer arm, ETPL (Exploit
Technologies Pte Ltd), during the Startup Asia Singa-
pore 2014 conference.
To realize this bug-repelling design concept,
Loh Xian Jun at the A*STAR Institute of Materials
Research and Engineering (IMRE) developed a spray-
able mixture of a fragrance oil and a polymer that emits
its scent at a sustained rate. The process is based on the
same microencapsulation technology used to control the
release of drugs.
Rather than tackle mosquitoes, the mixture was used
at the showcase to enhance the sensory experience of
several elegant dresses designed by students at Nanyang
Technological University in Singapore, under the
guidance of visiting assistant professor Galina Mihaleva.
Among the other wearable technologies enriching the
fashion collection were luminescent silks, printed elec-
tronic materials that switch from opaque to transparent
in response to changes in temperature, and sensor beads
that darken under extended exposure to ultraviolet light.
The purpose of the technology showcase, entitled
Next-to-the-Skin, was to bridge the gap between
research and product development, explains assistant
vice president at ETPL, Radiana Soh. We sought to
create a platform for scientists, fashion and industrial
designers, and industry professionals on which they
could engage and ideate at the early stages of commer-
cializing a technology.
By introducing the concept of design thinking in
the early stages of the technology transfer process, ETPL
was able to offer A*STAR researchers a new perspective
on developing products that address consumer needs
while providing innovators with a better understanding
of the technologies available to them.
The process was a game-changing experience for
me, reflects researcher Santiranjan Shannigrahi from
the IMRE, who created the ultraviolet-sensitive beads.
I would never have thought that our materials could
find a use in such applications.
The success of this new way of thinking was apparent
in the host of prototypes developed by A*STAR
researchers in the month leading up to the final exhibit.
For example, Serene Ng Lay Geok at the A*STAR
Data Storage Institute invented SleepPro, a smart
sleeping pillow that uses patented software developed
by the A*STAR Institute for Infocomm Research
to monitor the sleepers heart rate, breathing and
sleep patterns. The comfortable headrest can be used
by elderly or sick patients to alert caregivers of any
abnormal behavior.
Another prototype presented at the showcase was
MKool, a cooling pad designed by Shah Kwok Wei,
a researcher at the IMRE. The pad combines heat-
conductive nanofibers and a phase-change material that
stores heat by changing its physical state to keep the
bodys temperature at a comfortable level. Nanosilver
ions, also incorporated into the fabric, confer the pad
with antibacterial and odor-killing properties.
Beyond highlighting the potential for technological
innovation at A*STAR, Next-to-the-Skin also empha-
sized the commercial viability of prototypes such as
MKool, notes Soh. We hope to accelerate the process
of bringing our products to market by creating an
ecosystem for early-stage collaboration through events
such as this.
Researchers at A*STAR nd innovative ways of incorporating their technologies into
wearable fabrics and electronics
Technology on the catwalk
The revolutionary cooling fabric, MKool,
developed by Shah Kwok Wei at the IMRE
was incorporated into the white scarf
of the outfit , meaning air in Greek.
Designed by Derek Teong Shi Hong, is
the visual expression of a life-sustaining
breath of cool, pure air.

2
0
1
4

L
o
h

Q
i

Z
h
i

C
l
i
f
o
r
d

2
0
1
4

L
o
h

Q
i

Z
h
i

C
l
i
f
o
r
d

2
0
1
4

L
o
h

Q
i

Z
h
i

C
l
i
f
o
r
d
The elegant dress, Manjusaka, designed by
Zachery Cheong Zhan at Nanyang Techno-
logical University (NTU) was coated with a
special mixture developed by Loh Xian Jun
at the A*STAR Institute of Materials
Research and Engineering (IMRE) that
emits a continuously captivating fragrance.
IMRE researcher Han
Ming Yong and NTU
designer Kalaimathi
Mahendran wove
luminescent silk into
an arresting black
dress presented
at Startup Asia
Singapore 2014.
Research Highlights
CELL BIOLOGY &
IMMUNOLOGY
CELL BIOLOGY &
IMMUNOLOGY
Exosomes secreted from mesenchymal
stem cells help to increase the number
of regulatory T cells and prevent skin
graft rejection in mice.

E
u
g
e
n
e

L
l
a
c
u
n
a
/
i
S
t
o
c
k
/
T
h
i
n
k
s
t
o
c
k
Stem cells:
The secretions of
stem cells
Mesenchymal stem cells (MSCs)
can be extracted from many
different types of tissues and are
currently used in clinical trials for
a range of conditions, including
autoimmune diseases. Now, a team
of researchers led by Sai Kiang Lim
at the A*STAR Institute of Medical
Biology in Singapore has demon-
strated that small vesicles secreted
by MSCs, called exosomes, can
exert anti-inflammatory effects on
immune cells in tissue culture and
also in mice that have received
skin transplants
1
.
Exosomes contain a variety
of proteins and other factors
from their originating cells.
Previous studies have shown that
MSC-derived exosomes protect
heart tissue, prompting Lim and
colleagues to investigate whether
MSC exosomes could have an
effect on immune-cell function.
The researchers isolated MSC
exosomes but found that they did
not have a direct effect on the
proliferation of lymphocytes a
type of white blood cell taken
from the spleen. However, because
lymphocyte function is steered by
monocytes, another type of white
blood cell, the researchers decided
to further investigate whether MSC
exosomes could activate monocytes
instead. They discovered that MSC
exosomes reduced the expression
of pro-inflammatory factors and
increased the expression of anti-
inflammatory factors in monocytes.
Toll-like receptors on immune
cells are proteins that initiate
the immune response following
activation by many different factors,
including one found in MSC
exosomes called fibronectin 1
(FN1). By blocking FN1 with an
antibody, the researchers were
able to reduce the ability of MSC
exosomes to activate monocytes.
When Lim and colleagues
exposed the exosome-treated mono-
cytes to developing T cells, a type of
lymphocyte, the cells matured into
regulatory T (T
reg
) cells a cell
type that suppresses the immune
system. The findings suggest
that MSC exosomes probably act
directly on monocytes, which can
then modulate lymphocyte matura-
tion and function.
As a result of their immunosup-
pressive properties, T
reg
cells can
help to prevent the rejection of
skin grafts by the immune system.
To test whether MSC exosomes
could facilitate the process, the
researchers grafted skin onto
mice, and then treated some of the
grafted mice with MSC exosomes.
Rejection of the grafted skin
was delayed by a few days in the
exosome-treated mice compared
to normal mice, probably due to
the higher levels of T
reg
cells in the
exosome-treated mice.
Our findings suggest that MSC
exosomes could be used to alleviate
diseases that have a dysfunctional
immune component, such as lupus,
psoriasis and sepsis, explains Lim.
A*STAR is currently funding
the clinical development of MSC
exosomes, he says.
Tiny vesicles secreted by mesenchymal stem cells can modulate the
immune system and prevent the rejection of grafted tissue
1. Zhang, B., Yin, Y., Lai, R. C.,
Tan, S. S., Choo, A. B. H. & Lim, S. K.
Mesenchymal stem cells secrete
immunologically active exosomes.
Stem Cells and Development 23,
12331244 (2014).
CELL BIOLOGY &
IMMUNOLOGY
1. Chan, Y.-S., Gke, J., Ng, J.-H., Lu, X.,
Gonzales, K. A. U. et al. Induction
of a human pluripotent state with
distinct regulatory circuitry that
resembles preimplantation epiblast.
Cell Stem Cell 13, 663675 (2013).
Human embryonic stem cells
(hESCs) have the ability to both
convert into any cell type in the
human body and to proliferate
indefinitely in the laboratory.
However, cultured hESCs, which
are plucked from the developing
embryo and then grown in vitro,
often display a number of biological
differences when compared to the
pluripotent epiblast cells in the early
embryo from which they originated.
Now, a team of researchers at
the A*STAR Genome Institute of
Singapore (GIS) has developed a
way to make hESCs more closely
resemble true epiblast cells
1
.
We found that pluripotent
human embryonic stem cells can
be transformed into a state that is
closer to how they appear in real
life, says Huck Hui Ng, executive
director of the GIS, who led the
research. This new cell state has
the potential to improve many
applications of hESCs, such as
disease modeling.
To induce the more native state,
Ng and his colleagues screened
11 small molecules to find those
that promoted the expression of
NANOG, a gene involved with
self-renewal. The researchers identi-
fied a cocktail of three chemicals,
plus the growth factor LIF (3iL),
that nearly doubled NANOG
expression levels in the hESCs while
maintaining these cells in a state
of perpetual growth (see image).
They called the resulting stem cells
3iL hESCs.
The chemically treated and
non-treated hESCs expressed
similar levels of a handful of genes
that underlie pluripotency, the
ability to both self-renew and to
develop into any cell type. But
only the 3iL stem cells had a gene
expression profile that closely
matched that of epiblast cells taken
straight from early embryos. Study
author Jonathan Gke suggests
that laboratory culturing could
have rewired certain regulatory
networks in conventionally grown
hESCs. The 3iL treatment poten-
tially resets this process, he says.
Culture conditions have been
optimized to support self-renewal,
and these culture conditions are
very different from the complex
signaling and communication
system of embryos, notes
Yun Shen Chan, another author on
the study. Therefore, it seems like
the culture conditions used to grow
hESCs might partly explain the
differences between lab-cultured
hESCs and in vivo epiblast cells.
The researchers also explored
whether the 3iL recipe could be
applied to induced pluripotent stem
(iPS) cells, which are adult cells
reprogrammed into an embryonic-
like state. A full comparison
of 3iL iPS cells with standard
iPS cells still has to be done, Chan
says, but our preliminary results
indicate that 3iL could result in
iPS cells of higher quality.
We found that
pluripotent human
embryonic stem cells can
be transformed into a
state that is closer to how
they appear in real life.
Developmental biology:
Helping cultured stem cells get
back to nature
Cultured human embryonic stem cells given the 3iL treatment more closely
resemble natural stem cells from the developing embryo

2
0
1
4

A
*
S
T
A
R

G
e
n
o
m
e

I
n
s
t
i
t
u
t
e

o
f

S
i
n
g
a
p
o
r
e
Human embryonic stem cells treated with the 3iL recipe express higher levels of NANOG (green), a gene involved with the
self-renewal of cells.
CELL BIOLOGY &
IMMUNOLOGY
Recently, O-linked N-acetylglu-
cosamine (O-GlcNAc), a sugar
ring that reversibly modifies
proteins inside the cell nucleus
and cytoplasm in a process known
as O-GlcNAcylation, has been
revealed to be a key regulator of cell
signaling in the body. Researchers
believe that cellular changes
induced by O-GlcNAc may be
linked to chronic diseases, such as
diabetes and cancer. Character-
izing such dynamic relationships,
however, is challenging.
Normally, O-GlcNAc detec-
tion relies on shotgun mass
spectrometry (MS), which breaks
proteins into sequence-specific
peptide chains and measures their
molecular weight. But as O-Glc-
NAcylated peptides are present in
such small amounts, building up
concentrations sufficient for detec-
tion requires time-consuming and
tedious enrichment methods.
Now, Julien Maury and
Andre Choo from the A*STAR
Bioprocessing Technology Insti-
tute in Singapore and co-workers
have developed a simple way to
detect native O-GlcNAcylated
proteins even at 10,000-fold
dilution with a targeted MS
technique known as multiple
reaction monitoring (MRM)
1
.
Instead of scrutinizing all
possible peptides from a protein
system, MRM-MS filters out
masses that correspond to an
expected precursor ion. Then, the
joined-up precursor peptide and
selected fragment ions termed
transition couples are moni-
tored to build up a quantitative
assay of the protein structure.
Choo explains that by
focusing on specific O-GlcNAc-
modified targets, MRM-MS
enhances detection limits
without extensive labeling
or enrichment techniques.
Furthermore, he notes that the
methods ability to spot sugar
modifications in complex peptide
mixtures could greatly simplify
cell bioanalysis.
After programing the
MRM-MS system to spot standard
O-GlcNAcylated peptides, the
team turned their attention to the
enzyme glycogen synthase kinase-3
beta (GSK-3), which competes
with O-GlcNAc for binding sites on
proteins and is linked to numerous
high-profile diseases. Although
unconfirmed, GSK-3 may
itself be modified and regulated
by O-GlcNAc.
To investigate O-GlcNAcylated
GSK-3 peptides, the researchers
used gel electrophoresis to
extract GSK-3 from proteins
derived from human embryonic
stem cells (hESCs). Subsequent
MRM-MS analysis of the sample
revealed the presence of a novel
O-GlcNAcylated GSK-3 peptide
with three potential binding
sites a modification with
potential significance for the
enzymes autoinhibition mecha-
nism. Using their technique, the
researchers could also quantify
changes in O-GlcNAcylated
GSK-3 peptide levels following
hESC drug treatment.
Choo and Maury anticipate
that their work could help to detect
and quantify O-GlcNAcylated
peptides in samples where
protein amounts are scarce, such
as mouse brains. Additionally,
scientists want to know the
dynamic variations in O-GlcNAc
following drug treatment and cell
differentiation studies, they say.
Quantification by MRM-MS
is exactly suited to these sort
of investigations.
Protein samples ready for mass spectrometry (MS): multiple reaction monitoring MS
provides a reliable means to detect protein modification and may shed light on the
cellular mechanisms behind chronic diseases.
Bioanalysis:
Focusing on the sweet spot
1. Maury, J. J. P., Ng, D., Bi, X.,
Bardor, M. & Choo, A. B.-H.
Multiple reaction monitoring mass
spectrometry for the discovery
and quantification of O-GlcNAc-
modified proteins. Analytical
Chemistry 86, 395402 (2014).
A novel targeted mass spectrometry technique uncovers an elusive sugar
modication in stem cells

C
h
r
i
s
t
i
a
n

D
e
l
b
e
r
t
/
i
S
t
o
c
k
/
T
h
i
n
k
s
t
o
c
k
CELL BIOLOGY &
IMMUNOLOGY
Embryonic stem cells (ESCs)
have the potential to form more
than 200 distinct cell types
in the human body. Although
ESCs can differentiate into any
specialized tissue, scientists are
still unsure of how to coax these
pluripotent cells to reliably form
a desired cell type without pro-
ducing a mix of contaminating
cell lineages.
A research team led by
Bing Lim from the A*STAR
Genome Institute of Singapore
has now developed a new method
for directing ESCs into highly
pure populations of liver cells or
pancreas cells
1
. These untainted
populations of organ-specific
cells could form the basis of
future therapies or be used
as platforms upon which to
screen drugs.
Heterogeneous mixtures
of cell types are unsuitable for
transplantation or other thera-
peutic purposes, says
Kyle Loh, a previous member
of Lims lab now based at the
Stanford University School of
Medicine in the United States.
We can precisely differentiate
embryonic stem cells into a pure
population of a given lineage of
interest, he explains.
Lay Teng Ang of the liver
research program in Lims lab
sought a way to guide human
ESCs to differentiate reliably into
endoderm the cell type that
gives rise to organs including
the lungs, liver and intestines.
To do this, Ang systematically
perturbed developmental signals
at four consecutive steps of
endoderm formation, searching
for molecules that could produce
a single, desired cell type as
well as block the induction of
unwanted alternatives. Along the
way, the researchers generated
what Lim calls a roadmap for
endoderm differentiation.
Their strategy showed that
a variety of growth factors and
signaling proteins initially help
transform ESCs into endoderm
progenitor cells. However, the
researchers had to inhibit these
same molecules within 24 hours
to prevent the cells from turning
into another tissue type known
as mesoderm.
Knowledge of this timing
and the downstream signaling
dynamics eventually allowed
Lims team to differentiate the
ESCs into pure populations of
liver and pancreas cells, while
excluding other lineages at each
developmental branch point.
The researchers also noted the
endodermal enhancers existed
in a surprising diversity of
pre-enhancer states as uncom-
mitted cells prior to activation:
they also documented the per-
missive chromatin marks that
provide ESCs with their f lexible
developmental capacity.
We first needed to under-
stand the signals that control
stem cell differentiation and
thus what controls lineage
splitting, Loh says. Then
we could unilaterally repress
differentiation of ESCs toward
other cell types and instead drive
stem cells exclusively toward a
uniform population of desired
stem cells.
1. Loh, K. M., Ang, L. T., Zhang, J.,
Kumar, V., Ang, J. et al. Efficient
endoderm induction from human
pluripotent stem cells by logically
directing signals controlling lineage
bifurcations. Cell Stem Cell 14,
237252 (2014).
Embryology:
A roadmap for pure
organ cells
A new strategy minimizes the cellular contaminants that usually plague the
diferentiation of pluripotent stem cells

2
0
1
4

A
*
S
T
A
R

G
e
n
o
m
e

I
n
s
t
i
t
u
t
e

o
f

S
i
n
g
a
p
o
r
e
Liver cells derived from human pluripotent stem cells. The colors red and green mark the
expression of liver genes in the cells.
We can precisely
diferentiate embryonic
stem cells into a pure
population of a given
lineage of interest.
CELL BIOLOGY &
IMMUNOLOGY
Stem cells:
The fat source makes
the difference
Stem cells derived from diferent types of fat express diferent
cell-surface markers
Stem cells derived from distinct sources
of fat display different cell-surface
markers with implications for their
roles in metabolism and disease.

U
g
r
e
e
n
/
i
S
t
o
c
k
/
T
h
i
n
k
s
t
o
c
k
Mesenchymal stem cells (MSCs)
have a natural ability to differentiate
into various cell types, such as
muscle, cartilage and bone. They
can be classified according to
their source and include adipose-
derived stem cells (ASCs) and
bone-marrow-derived stem cells
(BMSCs). ASCs, in particular, hold
tremendous potential for tissue
engineering and regenerative medi-
cine because of their relatively high
abundance and ease of isolation.
Shigeki Sugii at the A*STAR
Singapore Bioimaging Consortium
and co-workers have now isolated
ASCs from two different sources of
fat: subcutaneous fat found under-
neath the skin and visceral fat from
inside the abdominal cavity
1
. The
team showed that ASCs derived
from subcutaneous fat express
cell-surface markers that differ from
those derived from visceral fat. The
finding has implications for deter-
mining the origins of ASCs and
the roles of their different subtypes
in metabolism-related conditions,
such as obesity, and diseases such as
soft-tissue tumors.
Stem cells, like all other cells,
express molecules at their surface
that are recognized by the bodys
immune system. Like a fingerprint,
the expression profile of these
cell-surface markers is unique
to each stem cell type. Scientists
already know that MSCs express
the cell-surface markers CD73,
CD90 and CD105 but not CD14,
CD19, CD34 and CD45. Recent
studies have also shown that
while ASCs express CD36 but not
CD106, the opposite is true for
BMSCs. Thus, MSCs derived from
different tissues express different
cell-surface markers, providing a
valuable tool for determining the
origins of MSCs.
Sugii and co-workers therefore
proposed that ASCs derived
from different types of fat may
also express different cell-surface
markers. To investigate this, they
extracted subcutaneous and visceral
fat from 12 obese patients, as well
as normal and obese mice. After iso-
lating ASCs from the fat, the team
analyzed the expression profiles of
over 240 cell-surface markers for
each sample. Their analysis revealed
a high level of CD10 expression in
ASCs derived from subcutaneous
fat compared to ASCs derived
from visceral fat. In addition, they
detected a high level of CD200
expression in ASCs derived from
visceral fat compared to those
derived from subcutaneous fat.
The researchers also discovered
that while CD10-rich ASCs from
subcutaneous fat differentiate better
than their CD10-deficient counter-
parts, CD200-deficient ASCs from
visceral fat differentiate better than
their CD200-rich counterparts.
Our results suggest that CD10
and CD200 are markers of high
and low adipogenic capacities,
says Sugii. Therefore, CD10 and
CD200 are biomarkers as well as
indicators of adipogenic potentials
for use in high-throughput drug-
screening systems.
1. Ong, W. K., Tan, C. S., Chan, K. L.,
Goesantoso, G. G., Chan, X. H. D. et al.
Identification of specific cell-surface
markers of adipose-derived stem
cells from subcutaneous and visceral
fat depots. Stem Cell Reports 2,
172179 (2014).
CELL BIOLOGY &
IMMUNOLOGY
The immunoglobulin E (IgE)
antibody released by the immune
system is a pivotal defense
against gut parasites and toxins.
However, the same antibody when
misdirected can also cause allergic
responses to food or substances in
the environment. An international
team led by researchers at the
A*STAR Singapore Immunology
Network has now discovered a
regulatory mechanism that keeps
IgE levels in check
1
.
A better understanding of the
process of IgE production and how
it can be subverted should enable
scientists to develop the next gen-
eration of allergy treatments. If we
know the mechanism, potentially
we could intervene therapeutically
to prevent the development of severe
allergic and anaphylactic reactions,
says A*STAR immunologist
Maria Curotto de Lafaille, who
led the research.
A type of white blood cell
known as a B cell can change
the antibodies that it produces
through a process known as
immunoglobulin class switching.
B cells that produce IgE can
arise through one of two distinct
switching mechanisms. Either the
B cells go through a sequential
process that involves another type
of immunoglobulin, IgG, as an
intermediary, or they take a direct
route to IgE generation. The
sequential process creates IgE with
high affinity for antigens the
substances that can trigger an
immune response and can cause
severe allergic reactions. In
contrast, the direct switching
pathway produces IgE antibodies
that bind their antigens weakly.
These low-affinity antibodies can
compete with the problematic IgE
to prevent anaphylaxis.
The origin, functional proper-
ties and population dynamics
of IgE-producing cells had been
poorly understood. To track the
dynamics of IgE production in a
living system, Curotto de Lafaille
and her colleagues created a new
mouse strain that fluoresces
whenever and wherever IgE
antibodies are produced. Using
the mouse model, the researchers
showed that IgE-producing cells
that undergo the direct switching
pathway at germinal centers
sites in the lymphatic system
where B cells differentiate and
mutate to enhance their antigen
binding tend to get stuck in
that developmental state and end
up failing to thrive. These cells
have a lot of defects and they end
up dying, says Curotto de Lafaille.
The failure of the IgE cells to
directly produce antibodies with
strong binding ability imposes a
strong constraint on IgE-based
immune responses. If you easily
make so many high-affinity IgE
cells, youd be constantly in danger
of life-threatening reactions like
anaphylaxis, Curotto de Lafaille
explains. Thats why theres an
evolutionary pressure to limit
this process.
Immunology:
Subverting the sneeze
Disease-causing IgE antibodies are kept under wraps to avoid
allergic responses
A delicate regulatory balance exists to keep IgE antibodies from causing allergic reactions.

a
n
t
p
k
r
/
i
S
t
o
c
k
/
T
h
i
n
k
s
t
o
c
k
1. He, J.-S., Meyer-Hermann, M.,
Xiangying, D., Zuan, L. Y.,
Jones, L. A. et al. The distinctive
germinal center phase of
IgE
+
B lymphocytes limits their
contribution to the classical
memory response. The Journal
of Experimental Medicine 210,
27552771 (2013).
CELL BIOLOGY &
IMMUNOLOGY
The binding of hormones to their
receptors plays a key role in the
development of many organs of
the body. The apelin receptor is
expressed in the developing embryo,
coming online many hours prior to
its presently known ligand, apelin.
Organisms that lack apelin have less
severe developmental defects than
those lacking the receptor, leading
scientists to believe that there is
an alternative ligand for the apelin
receptor that is expressed during the
very early stages of development.
Now, Bruno Reversade and col-
leagues at the A*STAR Institute of
Medical Biology and the A*STAR
Institute of Molecular and Cell
Biology in Singapore have identified
the hormone ELABELA as an
apelin-receptor ligand that is present
at the earliest stages of development.
The teams findings suggest that
ELABELA plays a key role in the
maturation of the endoderm and
the creation of the heart
1
.
Early embryonic development is
characterized by the development
of three germ layers: the ectoderm,
which forms skin and nervous
tissue; the mesoderm, which forms
the cardiovascular system; and the
endoderm, which forms the inside
of the gastrointestinal tract.
When the researchers gener-
ated zebrafish embryos lacking
ELABELA expression, they
found reduced levels of early
endodermal markers in the early
stage embryos. If they allowed the
embryos to grow to later devel-
opmental stages, Reversade and
colleagues found that the zebrafish
hearts either did not develop or
were highly abnormal. These
findings suggest that ELABELA
is required for the maturation of
endodermal cells that drive the
development of nearby meso-
dermal cells, which in turn go on
to form the heart.
As the abnormal heart
phenotype in ELABELA-deficient
zebrafish embryos was similar to
that of apelin-receptor-deficient
embryos, the researchers decided
to investigate if ELABELA acts
by binding to the apelin receptor
(see image).
They discovered that
ELABELA and the apelin receptor
are expressed at the same time
and similar location in the early
embryo. Expressing the apelin
receptor in cell lines that did
not normally express it allowed
binding of ELABELA to the
receptor at the cell surface, says
Reversade. These results suggest
that ELABELA is the earliest
ligand to bind to the apelin
receptor through the role it plays
in driving the primary stages of the
development of the embryo.
ELABELAs importance to
heart formation could also apply
to the development of other organ
systems. ELABELA is expressed
in kidneys and prostate in humans,
but its role there remains unknown
for now, notes Reversade.
Developmental biology:
Building a heart
Newly identied hormone ELABELA regulates the generation of the embryos
endoderm and is required for the normal formation of the heart
The ELABELA hormone has been identified as an important apelin-receptor ligand in
early heart development. Wild-type zebrafish embryos have normal cardiac develop-
ment (top), while embryos lacking the apelin receptor (center) develop similar cardiac
deformities to embryos lacking ELABELA (bottom).

2
0
1
3

E
l
s
e
v
i
e
r
1. Chng, S. C., Ho, L., Tian, J. &
Reversade, B. ELABELA: A hormone
essential for heart development
signals via the apelin receptor.
Developmental Cell 27,
672680 (2013).
CELL BIOLOGY &
IMMUNOLOGY
Immune cells known as
neutrophils are recruited by
chemical signals released from
sites of injury and infection. Their
primary purpose is to attack
pathogens and recruit additional
immune defenses, but these cells
can also inf lict tissue damage
through prolonged activation,
contributing to serious diseases
such as inf lammatory bowel
disease and chronic obstructive
pulmonary disorder.
Philip Ingham and colleagues
at the A*STAR Institute of
Molecular and Cell Biology
in Singapore have developed a
fish-based screening technique
that can rapidly identify com-
pounds to potentially control such
excessive neutrophil activation
1
.
Previously, Ingham collaborated
with Stephen Renshaws team at
the University of Sheffield, United
Kingdom, to generate a geneti-
cally modified zebrafish strain in
which neutrophils are selectively
labeled through production of a
f luorescent protein
2
.
The zebrafish model allowed
for easy observation of neutrophil
behavior in the transparent living
zebrafish embryos and larvae. We
showed that neutrophil migration
could be blocked by exposing fish
larvae to known migration inhibi-
tors, explains Ingham. Following
on these findings, Ingham and
Renshaw adapted the model to
screen a library of fungal extracts
and identify new molecules with
similar migration-inhibiting effects.
Most chemical screens in
zebrafish use pure compounds of
known chemical composition,
says Ingham. We were able to
discover compounds with specific
biological effects by screening
fairly complex mixtures found in
natural extracts.
Starting with over
1,000 extracts produced by
Singapore-based company
MerLion Pharmaceuticals, the
researchers identified two extracts
that inhibited neutrophil recruit-
ment in zebrafish following tail
injury. Fractionation of the two
revealed the active ingredients
responsible for the effect. Since
one of the molecules had previ-
ously been linked to toxic effects,
Ingham and colleagues focused on
the other molecule, PF1052.
Notably, they found that
PF1052 specifically blocks
neutrophil migration to zebrafish
tail wounds without affecting
other immune cells (see image).
They also found that the extract
appears to block formation of the
structures that coordinate cellular
movement. It does this by inter-
fering with the process of cellular
polarization the mechanism by
which cells establish their relative
orientation. Interestingly, PF1052
does not affect polarization via
established mechanisms, which,
Ingham suggests, shows there is
a novel route to controlling the
inf lammatory response.
Although the research effort
was primarily intended as a proof-
of-concept study, Ingham believes
that it opens up wider possibilities
and that the effects of PF1052
merit further investigation. The
next step would be to test PF1052
in an in vivo mammalian inf lam-
mation model to verify its efficacy
and at the same time assess its
potential toxicity, he says.
Immunology:
Fishing for new
therapeutics
A zebrash-based screening method reveals compounds that could help
contain the efects of immunological disorders
Neutrophils (green) normally migrate to the site of a tail wound (white line) in animals subjected to a control treatment (top). In
comparison, the fungal compound PF1052 inhibits neutrophil migration (bottom).
R
e
p
r
o
d
u
c
e
d

f
r
o
m

R
e
f
.
1

a
n
d

l
i
c
e
n
s
e
d

u
n
d
e
r

C
C

B
Y

3
.0

(
c
r
e
a
t
i
v
e
c
o
m
m
o
n
s
.o
r
g
/
l
i
c
e
n
s
e
s
/
b
y
/
3
.0
/
l
e
g
a
l
c
o
d
e
)

2
0
1
4

X
.
W
a
n
g

e
t

a
l
.
1. Wang, X., Robertson, A. L., Li, J.,
Chai, R. J., Haishan, W. et al.
Inhibitors of neutrophil recruitment
identified using transgenic
zebrafish to screen a natural
product library. Disease Models &
Mechanisms 7, 163169 (2014).
2. Renshaw, S. A., Loynes, C. A.,
Trushell, D. M. I., Elworthy, S.,
Ingham, P. W. & Whyte, M. K. B.
A transgenic zebrafish model of
neutrophilic inflammation. Blood
108, 39763978 (2006).
CELL BIOLOGY &
IMMUNOLOGY
Red blood cells are released into
the blood stream in their immature
form reticulocytes from the
bone marrow where they develop.
Reticulocytes are important markers
for certain blood disorders and
infectious diseases but their matura-
tion has been poorly understood.
Now, an international research
team led by Laurent Renia from the
A*STAR Singapore Immunology
Network has characterized, in fine
detail, the properties of reticulocytes
at different stages of maturation
1
.
Studies in the 1930s identified
distinct subtypes of reticulocytes
but only provided basic descrip-
tions of their different maturation
stages. According to Renia, these
early studies have mostly been
forgotten, and more detail is
needed to better understand the
reticulocyte maturation process.
The quantity and type of
reticulocytes released into the
circulation provide important
information for the diagnosis and
prognosis of certain diseases,
explains Renia. Despite the
significance of reticulocytes,
limited information is available
about their biology and it is
incorrectly assumed that reticulo-
cytes are a fairly uniform type of
blood cell.
Aided by recent advances in
f luorescent staining, Renia and
his team set out to describe the
properties of reticulocytes at
different stages of maturation
by studying cells from the
reticulocyte-rich blood of the
human umbilical cord. They
separated the reticulocytes into
four subpopulations according
to the level of their expression of
a specific surface protein, which
decreases as the cells mature.
This helped us to fully
characterize the chemical and
biomechanical differences of these
reticulocyte stages, says Renia.
They found that as reticulocytes
mature, the outer membrane
and internal structures are
reorganized, creating cells that
are smaller, less rigid and have
the characteristic concave shape
of red blood cells (see image).
They also revealed changes
in expression levels of specific
proteins and the attenuation of
metabolic pathways.
We uncovered an incredible
level of reticulocyte heterogeneity
characterized by significant
chemical, biophysical and meta-
bolic modifications, says Renia.
No-one else has characterized
reticulocytes on such a fine scale.
The differences the team noted
demonstrate that reticulocyte
maturation is a continual process,
meaning that reticulocytes
should not be grouped together
as one cell type. In addition, the
finding hints that the matura-
tion process could have direct
clinical applications.
Reticulocytes are important
in the study of vivax or relapsing
malaria, the most common
form of malaria in Asia, because
the Plasmodium vivax parasite
specifically targets reticulocytes,
he says. We hope our findings
will help to develop vaccines that
block the blood stage life cycle of
vivax malaria parasites.
Cell biology:
Detailing the development of
red blood cells
Understanding the diversity of immature red blood cells in greater detail
could help protect against Asias most common form of malaria
Reticulocyte maturation
Red blood cell
The four subpopulations of reticulocytes show altered external appearances (top) and
internal structures (bottom) as they mature into red blood cells.
M
o
d
i
e
d

f
r
o
m

R
e
f
.
1

a
n
d

l
i
c
e
n
s
e
d

u
n
d
e
r

C
C

B
Y

3
.0

(
c
r
e
a
t
i
v
e
c
o
m
m
o
n
s
.o
r
g
/
l
i
c
e
n
s
e
s
/
b
y
/
3
.0
/
l
e
g
a
l
c
o
d
e
)

2
0
1
3

B
.
M
a
l
l
e
r
e
t

e
t

a
l
.
1. Malleret, B., Xu, F., Mohandas, N.,
Suwanarusk, R., Chu, C. et al.
Significant biochemical, biophysical
and metabolic diversity in
circulating human cord blood
reticulocytes. PLoS ONE 8,
e76062 (2013).
We hope our ndings
will help to develop
vaccines that block the
blood stage life cycle of
vivax malaria parasites.
No-one else has
characterized
reticulocytes on such
a ne scale.
CELL BIOLOGY &
IMMUNOLOGY
City life can be an assault on
the senses quite literally, in
the case of allergies. The steady
increase in global urbanization
is mirrored by a growing preva-
lence of allergy-associated respira-
tory problems. Potential triggers
include insects, mold, pollen and
animal hair, but now A*STAR
researchers have uncovered a single
culprit with a disproportionate
role in allergy onset in tropical
urban settings
1
.
Allergies arise when the body
mounts an immune response to a
foreign molecule that it mistakenly
perceives as a threat. Whenever the
body encounters that trigger for
example, a pollen grain it
produces large numbers of
antibodies against the trigger. The
resulting inflammatory response
can cause symptoms including
asthma and rhinitis. Olaf Rotzschke
and colleagues from the Singapore
Immunology Network and
De Yun Wang of the National
University of Singapore began their
study by surveying the antibody
responses of 206 volunteers to a
dozen common allergic triggers.
Remarkably, the great majority
of these Singapore-born individuals
responded to one particular
antigen: the tiny dust mites found
in many homes (see image). This
trend remained clear even after
the researchers expanded their
cohort to look at a larger group
of individuals. According to our
study, 80 per cent of Singaporeans
respond to the mite, with roughly
40 per cent developing allergic
rhinitis and 15 per cent developing
asthma, says Rotzschke. Globally
these are among the highest figures
reported so far.
The allergic reaction appears
to be a consequence of the Singa-
porean urban environment. When
the researchers examined newly
arrived Chinese immigrants, fewer
than 30 per cent mounted a strong
antibody response against dust
mites; in contrast, for immigrants
who had lived in Singapore eight
years or longer, the response rate
was indistinguishable from lifelong
Singapore residents. This phenom-
enon of gradual acquisition of an
allergic reaction has been shown in
other countries for other allergens as
well, explains Rotzschke. However,
in temperate regions in Western
countries, the most common allergic
trigger is pollen, suggesting that this
dust-mite-associated sensitization
may be more characteristic of
tropical cities like Singapore.
By identifying a single target,
the findings could be used to
provide relief to large numbers of
allergy sufferers. But the existence
of such a large population with a
shared, strong response to a single
antigen has broader implications for
research as well, notes Rotzschke.
We are currently planning a func-
tional analysis of immune pathways
in combination with genome-wide
genetic studies to better characterize
the molecular and genetic basis of
allergies, he says.
Immunology:
Putting the squeeze on the
Singapore sneeze
House dust mites may be the primary trigger for allergy-associated
respiratory problems in tropical urban environments such as Singapore
The house dust mite, a common but invisible presence in many homes, is a major trigger of allergic respiratory problems in Singapore.

E
r
a
x
i
o
n
/
i
S
t
o
c
k
/
T
h
i
n
k
s
t
o
c
k
1. Andiappan, A. K., Puan, K. J., Lee, B.,
Nardin, A., Poidinger, M. et al. Allergic
airway diseases in a tropical urban
environment are driven by dominant
mono-specific sensitization against
house dust mites. Allergy 69,
501509 (2014).
CELL BIOLOGY &
IMMUNOLOGY
Human stem cells produced
through genetic reprogramming
are beset by safety concerns because
current techniques alter the DNA
of the stem cells and use material
from animals to grow them. Now,
A*STAR researchers have developed
an efficient approach that produces
safe, patient-specific human
stem cells
1
.
Human induced pluripotent
stem cells have the potential to
treat a number of diseases without
the ethical issues associated with
embryonic stem cells. Pluripotent
stem cells can be produced from
adult cells by introducing genes
that reprogram them. Typically,
the stem cells are grown on a
layer of mouse cells in solutions
(known as media) that contain
animal proteins and therefore,
potentially may also carry disease.
For such stem cells to be safe for
use in humans, they need to be
grown in xeno-free conditions,
which are devoid of material from
other animals.
Andrew Wan and Hong Fang
Lu at the A*STAR Institute of
Bioengineering and Nanotech-
nology in Singapore and colleagues
set out to develop a new xeno-free
system. The researchers carried
out the genetic reprogramming of
cells on an artificially produced
protein substrate rather than mouse
cells. They also used media that
contained no animal components.
The result was more efficient
reprogramming than seen with
conventional approaches.
A xeno-free system will
eliminate the risk of disease trans-
mission from other species, which is
important for regulatory approval,
explains Wan. Yet there have been
few studies on cell reprogramming
under totally xeno-free conditions.
The researchers went one step
further by addressing the problem
of cells acquiring alterations to their
DNA during reprogramming.
Incorporation of transgenes into
the genome of the cell poses another
safety issue, risking unwanted
genetic alterations, explains Lu.
In our work, the transgenes
were introduced to initiate the
reprogramming, but after this they
were removed from the cell, leading
to transgene-free stem cells.
The researchers demonstrated
that after genetic reprogramming
and the removal of the added genes,
the stem cells could still develop
into different cells types. They
were even able to induce them
to form dopaminergic neurons,
the type that degenerates in
Parkinsons disease. The conditions
in which the stem cells were grown
mean that they are suitable for
clinical use and can be derived
from a patients own cells, ensuring
complete compatibility.
Regulatory approval for clinical
application of stem cells largely
depends on the conditions in
which the stem cells are derived,
says Wan. We present a workable
protocol for the reprogramming of
fibroblasts to stem cells that mini-
mizes any potential safety risks.
Stem cells:
Animal-free reprogramming
improves safety
Reprogramming of adult cells in animal-free conditions provides a safer
culture system for therapeutic stem cells
Growing stem cells in conditions free of
animal material makes them safe for use
in humans.

E
r
a
x
i
o
n
/
i
S
t
o
c
k
/
T
h
i
n
k
s
t
o
c
k
1. Lu, H. F., Chai, C., Lim, T. C.,
Leong, M. F., Lim, J. K. et al.
A defined xeno-free and feeder-
free culture system for the
derivation, expansion and direct
differentiation of transgene-
free patient-specific induced
pluripotent stem cells. Biomaterials
35, 28162826 (2014).
CELL BIOLOGY &
IMMUNOLOGY
The human body maintains
a healthy layer of skin thanks
to a population of stem cells
that reside in the epidermis.
Previously, the signals responsible
for regulating these so-called
interfollicular epidermal stem
cells (IFESCs) were unclear, but
now scientists in Singapore and
the United States have shown
that these cells secrete proteins
in the Wnt signaling pathway to
control their own balance between
renewal and differentiation
1
.
The mechanism of stem
cell function that we describe is
novel to the skin and, as far as
we know, also unprecedented in
the entire stem cell field, says
Xinhong Lim of the A*STAR
Institute of Medical Biology, who
led the study. Skin stem cells
themselves can be a source of
stem cell self-renewing signals.
This expands the paradigm
for how mammalian stem cells
are regulated.
Lim and his colleagues focused
on Wnt proteins because of
their well-established roles in
stem cell maintenance and hair
growth. To determine if IFESCs
respond to Wnt signals, the
researchers created transgenic
mice in which they could visually
track the fate of cells producing
Axin2, a protein triggered by
Wnt activity.
They discovered that Axin2
was expressed by cells that
contribute to wound healing,
found in the deepest layer of the
epidermis (see image). Through
detailed molecular analysis, the
team confirmed that these cells
were IFESCs. In addition, they
noted that the same stem cells
also expressed several Wnt family
genes themselves. Together, these
findings indicate that IFESCs are
both the source and the target of
Wnt signals.
The researchers concluded
that the self-renewal of IFESCs
is maintained by an autocrine
or self-stimulating loop of Wnt
production. The stem cells,
rather than a separate stem cell
niche, regulate epidermal thick-
ness and regeneration, Lim says.
These results led the team to
question how the IFESCs could
escape this regulatory loop and
commence the differentiation
process. Seeking answers, Lim and
his colleagues tracked the expres-
sion patterns of Wnt inhibitors in
the skin.
They found that while the
stem cells also secreted Wnt
inhibitors, these signals localized
to the cells situated above the stem
cell compartment, reinforcing dif-
ferentiation there. By producing
both an autocrine Wnt signal and
a paracrine Wnt inhibitor, Lim
explains, the stem cells in the
skin act as the organizing center
for the tissue.
The teams discovery could
lead to new therapies that manip-
ulate Wnt signaling to improve
wound healing. According to
Lim, it could advance skin stem
cell culture protocols, helping
scientists to grow skin grafts
more efficiently.
Stem cells:
Skin cells control their
own fate
The balance of renewal and diferentiation of the skins stem cells is
regulated by factors secreted by the cells themselves
Axin2 (red) and Wnt (turquoise) mRNA transcripts are expressed by stem cells at the
base of the mouse epidermis. Differentiated cells lie close to the surface of the skin.
R
e
p
r
o
d
u
c
e
d

f
r
o
m

R
e
f
.
1

2
0
1
3

X
.
L
i
m

e
t

a
l
.
1. Lim, X., Tan, S. H., Koh, W. L. C.,
Chau, R. M. W., Yan, K. S. et al.
Interfollicular epidermal stem
cells self-renew via autocrine Wnt
signaling. Science 342,
12261230 (2013).
CELL BIOLOGY &
IMMUNOLOGY
The upper respiratory tract is the
first line of defense against air
pollutants, including allergens,
bacteria and environmental
toxicants. Finger-like protrusions
called cilia on the surface of the
human mucous membrane, or
epithelium, sway back and forth
when irritated. This coordinated
beating movement of the cilia
helps to remove foreign materials
and is an important protec-
tive mechanism.
Wei Wang and Zhi Ping
Wang at the A*STAR Singapore
Institute of Manufacturing
Technology, De Yun Wang at
the National University of Sin-
gapore and co-workers have now
developed the first microf luidic
device that enables the direct
observation of cilia and their
beating frequency on a polyester
membrane
1
. The artificial system
is used to observe the effects
of air pollutants on cells in the
upper airway.
The researchers constructed
their microfluidic device using
glass and a transparent, moldable
polymer to ensure a clear view
of the cilia and their activity. A
membrane designed to support
the cultivation and differentiation
of human nasal epithelial stem
cells was inserted into a small
chamber on the device and fresh or
contaminated air was fed through
a tiny channel.
Five weeks after seeding the
membrane with human nasal epi-
thelial stem cells the researchers
could observe the formation of
beating cilia. The cilia beating
frequency varied across samples
but the difference was typically
within a few hertz.
When fresh air was passed
into the chamber, the researchers
observed a 3 per cent drop in cilia
beating frequency relative to the
baseline value. In contrast, when
air mixed with 0.5 milligrams per
cubic meter of formaldehyde was
passed through the chamber, they
observed a 7.4 per cent increase in
cilia beating frequency relative to
the baseline value.
Further increasing the
formaldehyde concentration to
1 milligram per cubic meter led
to a dramatic increase up
to 136.4 per cent in cilia
beating frequency relative to the
baseline value. At even higher
formaldehyde concentrations of
3.0 milligrams per cubic meter,
however, the researchers observed
an unexpected decrease in cilia
beating frequency, possibly due to
irreversible cell damage caused by
the formaldehyde.
By enabling the observation
of cilia beating frequency, the
experimental model described in
this work provides more realism
for clinical applications. The
device can be used to directly test
for toxicity and toxic mechanisms,
screen for drugs that reduce
irritation, and assess the level of
risks associated with a particular
air pollutant. The technology has
applications in chemical analysis,
environmental monitoring,
medical diagnostics and cellular
studies, says Wang.
Seeding a membrane on a microfluidic chip with human nasal epithelial stem cells
enables the formation and observation of beating cilia.
Microuidics:
Lab on a breathing chip
1. Wang, W., Yan, Y., Li, C. W., Xia, H. M.,
Chao, S. S., Wang, D. Y. & Wang, Z. P.
Live human nasal epithelial cells
(hNECs) on chip for in vitro testing
of gaseous formaldehyde toxicity
via airway delivery. Lab on a Chip
14, 677680 (2014).
Human nasal epithelial cells, cultured on a microchip, react to air pollutants
just like they would in the upper airway

D
A
J
/
T
h
i
n
k
s
t
o
c
k
Research Highlights
CHEMISTRY &
MATERIALS
1 m
CHEMISTRY &
MATERIALS
A protocol that improves the way that
data is written to a solid-state drive
improves device efficiency and lifetime.

O
l
e
k
s
i
y

M
a
r
k
/
i
S
t
o
c
k
/
T
h
i
n
k
s
t
o
c
k
Data storage:
Ensuring solid-state drives are up
to scratch
Solid-state drives (SSDs) store
digital information using electronic
circuits. The power efficiency of
SSDs and their ability to read and
write data quickly means that
they are becoming the primary
storage device in computers. A
major drawback of SSDs, however,
is the limited number of times
that data can be stored and
deleted an aspect that hinders
the use of these devices for data-
intensive applications known as
data-center environments.
Qingsong Wei and co-workers
at the A*STAR Data Storage
Institute and National University of
Singapore have developed a scheme
for writing data to SSDs that could
circumvent these problems to make
solid-state drives useful for an even
broader range of applications
1
.
SSDs divide their storage space
into distinct areas called blocks.
A computer can either save large
files across consecutive blocks a
process known as sequential
writing or write smaller files in
blocks scattered throughout the
device so-called random writing.
The researchers conducted
an intensive workload study of
the distribution of read and write
request sizes over ten real enterprise
workload traces supplied by the
Storage Network Industry Associa-
tion. They found that the highest
traffic was from small, random
requests of less than 64 kilobytes
in size.
Generally, random writing is
much slower by as much as four
times than sequential writing.
One way around this bottleneck
is to use part of the memory as a
buffer. The buffer briefly stores
data as it comes into the drive,
which then enables sequential
writing at a later time. Current
buffer management approaches
improve sequential writing but
only at low buffer usage, wasting
expensive buffer space.
Weis team helped to solve this
problem through an alternative
approach that categorizes the data
in the buffer by its popularity,
which reflects how frequently the
data is likely to be needed. The
scheme retains popular blocks in
the buffer, rather than deleting
them, and sequentially writes less
popular blocks to the SSD.
Our buffer management
scheme can increase sequential
writing with high buffer utilization,
thus improving performance and
extending the lifetime of the SSD,
says Wei.
The researchers tested the
approach and demonstrated that
the so-called popularity-aware
buffer management scheme, or
PAB, can achieve an improvement
in performance of up to 72 per cent
and triple the device lifetime
compared to existing schemes. Our
method reduces the cost of SSDs by
improving buffer utilization and is
easy to implement, explains Wei.
Our next step will be to design
smarter SSDs by integrating these
same ideas with emerging non-
volatile memory.
A data bufering scheme improves the performance of solid-state drives in
large-scale, data-intensive applications
1. Wei, Q., Zeng, L., Chen, J. &
Chen, C. A popularity-aware buffer
management to improve buffer
hit ratio and write sequentiality
for solid-state drive. IEEE
Transactions on Magnetics 49,
27862793 (2013).
CHEMISTRY &
MATERIALS
1. Zhang, L., Oh, S. R., Wong, T. C.,
Tan, C. Y. & Yao, K. Piezoelectric
polymer multilayer on flexible
substrate for energy harvesting.
IEEE Transactions on Ultrasonics,
Ferroelectrics and Frequency
Control 60, 20132020 (2013).
The shrinking dimensions and
decreased power consumption
of modern electronic gadgets
have created opportunities for
energy harvesting processes that
tap into free, green energy from
the environment. Vibration
harvesters, for example, produce
small amounts of electricity from
everyday mechanical disturbances
such as wind currents, traffic noise
or footsteps.
Now, Kui Yao and co-workers
from the A*STAR Institute of
Materials Research and Engineering
in Singapore have discovered a
way to give lightweight polymer
vibration harvesters a hundredfold
boost in energy output a finding
that may help to eliminate manual
battery recharging in microsensors
and mobile devices
1
.
Many vibration harvesters
contain piezoelectric substances
that create an electric voltage
when mechanically bent. By
fabricating piezoelectric materials
into cantilevers that resemble a
diving board, these devices can
oscillate from ambient vibrations
and generate electricity. Researchers
often use piezoelectric ceramics
because they impart large amounts
of electrical charges; however, the
brittleness of ceramics makes them
unsuitable for prolonged and large
vibrational movements.
Yao and co-workers investigated
a plastic-based piezoelectric
material, polyvinylidene fluoride
(PVDF), which is low cost and
readily undergoes mechanical
strain. To make efficient vibration
harvesters from PVDF, researchers
must stack the polymer in multiple
layers, improving the output
current and reducing the electrical
impedance that is inherent to
piezoelectric materials. But when
too many thin piezoelectric layers
are stacked, the cantilever can
become too stiff for bending-mode
vibrational harvesting.
To optimize piezoelectric har-
vesting with plastic films, the team
deployed an analytical approach.
Developing a mathematical model
of a multilayered polymer cantilever
coated with metal electrodes, the
researchers systematically calculated
how different material parameters
affected the energy output.
Their simulations revealed
some often-ignored factors such
as the thinness of electrode coat-
ings and the materials electrical
parameters, says Yao. These
can have a dramatic effect on the
electricity generated by bending
multilayer polymers.
One key parameter identified
was the need to match the electrical
impedance with an optimum load
resistance. The researchers analysis
showed that the energy output of a
22-layered piezoelectric structure
could be from 5 to 400 times higher
than a single-layer piezoelectric
polymer of similar dimensions.
The team then tested the
feasibility of their analytical results
by fabricating a PVDF-based
vibrational harvester on a flexible
aluminum substrate. They used
scalable dip-coating procedures
to build up polymer multilayers
and ensured thin metal electrode
coatings with physical vapor
deposition techniques.
Our experimental results are
promising and show that, for many
practical applications, piezoelectric
polymer multilayers may enable
harvested energy to replace bat-
teries, notes Yao.
Energy harvesting:
Multilayer polymers spring
into action
Flexible plastics that turn mechanical vibrations into electrical energy could
spur the development of self-powered sensors and devices
A green alternative to batteries that could power electronic devices is one step closer thanks to multilayered polymer cantilevers that
can turn mechanical vibrations into electricity.

E
y
e
m
a
t
r
i
x
/
i
S
t
o
c
k
/
T
h
i
n
k
s
t
o
c
k

CHEMISTRY &
MATERIALS
Ferroelectric materials have an
intrinsic electrical polarization
caused by a small shift in the
position of some of their atoms
that occurs below a critical point
called the Curie temperature.
This polarization can be switched
by an external electric field, an
effect exploited in some computer
memory devices.
By explaining the origin of
puzzling polarization patterns previ-
ously seen in a ferroelectric material
called barium titanate, Rajeev
Ahluwalia and Nathaniel Ng at the
A*STAR Institute of High Perfor-
mance Computing in Singapore
and colleagues have stumbled on a
way to write polarization patterns
in nanoscale ferroelectric materials
1
.
Ferroelectric crystals contain
a patchwork of nanoscale
domains, each with a different
intrinsic polarization. While
an understanding of how these
domains form would help to
develop reliable applications for
ferroelectric materials, two different
imaging techniques previously
revealed contradictory results about
the domains in barium titanate.
Ahluwalias team therefore set out
to solve this puzzle.
One technique transmission
electron microscopy (TEM)
which uses a beam of electrons to
probe a crystals properties, suggests
that the domains comprise long strips
arranged in four quadrants, where
the net polarization in each quadrant
points inward or outward from
the surface. The other technique
piezoresponse force microscopy
(PFM) also reveals a quadrant
formation, but the polarizations are
parallel to the surface so that the
overall polarization of the crystal
forms a closed loop (see image).
Ahluwalia and his colleagues
hypothesized that the TEMs
electron beam changes the polariza-
tion pattern in the sample. PFM, in
contrast, uses a sharp tip to detect
deformations in the material caused
by a localized electric field.
The scientists developed a
theoretical model, which revealed
that an increase in electron density
in the crystal produced the same
polarization pattern that they
observed with TEM. They also
calculated that the radial electric
field created by an electron beam
could generate other distinctive
features of this pattern.
Under normal conditions, an
electron beam might not alter the
domains. But if the beam is strong
enough to heat the sample above
the Curie temperature, the material
loses its intrinsic polarization. As
it cools, the radial electric field
induced by the electron beam
shapes how the domains reform.
The teams discovery serves as a
warning that electron beam tech-
niques could alter the very domains
that researchers are seeking to
measure. However, electron beams
could be used to deliberately alter
polarization patterns in ferroelectric
materials, something that is poten-
tially useful for the next generation
of memory devices with higher
storage densities, says Ahluwalia.
A simulated polarization pattern (top left), polarization vectors within the simulation
(top right), and polarization patterns visualized using transmission electron microscopy
(bottom left) and piezoresponse force microscopy (bottom right) in the ferroelectric
material barium titanate.
Materials:
Probing polarization
puzzles
1. Ahluwalia, R., Ng, N., Schilling, A.,
McQuaid, R. G. P., Evans, D. M. et al.
Manipulating ferroelectric domains
in nanostructures under electron
beams. Physical Review Letters 111,
165702 (2013).
Using electron beams to encode data in nanocrystals could help to improve
the capacity of computer memory devices
R
e
p
r
o
d
u
c
e
d

f
r
o
m

R
e
f
.
1

2
0
1
4

A
m
e
r
i
c
a
n

P
h
y
s
i
c
a
l

S
o
c
i
e
t
y
CHEMISTRY &
MATERIALS
Trapping light into a small volume
is a useful way of amplifying
optical effects. Optical cavities,
for example, can enhance the
interaction between light and
matter. Incorporating these tiny
structures into actual devices is
difficult however, because they
are easily broken or can become
optically misaligned.
Xia Yu at the A*STAR Singapore
Institute of Manufacturing Tech-
nology and co-workers have now
developed an optical-fiber-based
structure that harnesses the poten-
tial of light trapped in a microcavity.
The novel design also provides a
robust route to advanced devices for
filtering and sensing light
1
.
Yu and colleagues melted
silica glass to form a sphere with
a diameter of 182 micrometers.
They then patterned the end
of an optical fiber with a gold
grating and held it close to the
microsphere. The grating coupled
light propagating along the fiber
into the sphere (see image). Light
with the right wavelength traveled
around in circles within the sphere,
trapped by the smooth silicaair
interface. This confined light is
known as a whispering-gallery
resonant mode.
The A*STAR team investigated
the properties of their structure by
measuring the amount of light at
each wavelength that managed to
escape from the cavity back into
the fiber. The typical wavelength-
dependent response of a
microsphere is a sharp, symmetric
peak centered on the resonant
wavelength of the cavity.
Instead, the researchers
observed an asymmetric spectral
peak, which they recognized as
a clear signature of the so-called
Fano effect, indicating strong
interaction or interference
between the whispering-gallery
mode and the light in the fiber
directly ref lected back from
the grating.
This interfering effect makes
Fano resonances especially
sensitive to changes in either
of the participating systems: a
slight perturbation results in
dramatic alteration in the optical
characteristics, says Yu. An
obvious application of Fano
resonance is for use in ultra-
sensitive detection.
In previous investigations
of the optical Fano effect,
researchers inserted (and
extracted) light into the cavity
through the side of an optical
fiber an approach that
proved unstable and inefficient.
The method used by Yu and
colleagues of directly inserting
light into the cavity through the
end of the fiber proved far more
robust, making the technology a
plausible platform for cheap and
compact optical-resonator-based
photonic devices.
Another possible applica-
tion for the technology is as
an optical switch. A good
switching device must be fast,
explains Yu. Therefore, the
next step in our research will be
to attempt to control the speed
of the whispering-gallery mode
Fano resonance.
1. Zhou, Y., Zhu, D., Yu, X., Ding, W. &
Luan, F. Fano resonances in metallic
grating coupled whispering gallery
mode resonator. Applied Physics
Letters 103, 151108 (2013).
Optics:
The direct approach to
microcavities
A robust micrometer-scale structure for trapping light enhances optical
interactions in advanced photonic devices
A gold grating at the end of an optical fiber couples light into a spherical microcavity,
shown here as a purple sphere. The red arrows show the light bent by the grating,
which is then coupled to the whispering-gallery mode (red circle). Concentrating light in
compact structures is important for advancing photonics.
The next step in our
research will be to
attempt to control the
speed of the whispering-
gallery mode Fano
resonance.

2
0
1
4

A
*
S
T
A
R

S
i
n
g
a
p
o
r
e

I
n
s
t
i
t
u
t
e

o
f

M
a
n
u
f
a
c
t
u
r
i
n
g

T
e
c
h
n
o
l
o
g
y

CHEMISTRY &
MATERIALS
Bioananalysis:
Microbeads are
easily fixed
A passive method for sorting and xing microbeads of diferent sizes could
lead to cheaper and more functional biological assays
Microscopy image of a microtrap
array showing the capture of large
16-micrometer-diameter beads in
surface traps. Smaller 8-micrometer-
diameter beads accumulate in the
underlying dispersion gap.

2
0
1
4

A
*
S
T
A
R

I
n
s
t
i
t
u
t
e

o
f

M
i
c
r
o
e
l
e
c
t
r
o
n
i
c
s
Biological assays are an integral
part of the researchers toolkit in
the fields of biomolecular chemistry
and genomics. Microfluidic
microbead systems, which consist
of arrays of beads coated with an
assay-specific reagent, have revolu-
tionized biological assay technology
by allowing the high-throughput
detection of target molecules from
small sample volumes. Fabrication
of the microbead systems, however,
requires great care and various
ancillary devices.
Chee Chung Wong and col-
leagues from the A*STAR Institute
of Microelectronics have now
developed a passive and robust
method for manufacturing sorted
arrays of multiple microbead types
1
.
The preparation of microbead
systems conventionally involves
the use of a pump to introduce
a bead-carrying fluid into a
microfluidic circuit. The beads then
adsorb to the walls of the micro-
channels with little control over
position or sorting. The resultant
microbead-coated channels can be
used for targeted molecule detec-
tion, but the beads can be easily
dislodged by flow.
Recognizing the limitations of
conventional systems, Wong and
his colleagues set out to develop
a passive, pumpless method for
preparing more robust microbead
arrays. There are no pumpless bead
sorting strategies currently avail-
able, notes Wong. As a result, we
had to research and study three-
dimensional trap architectures that
could efficiently perform size-based
bead sorting.
The researchers used semicon-
ductor fabrication technologies
to create a trap architecture
consisting of a top surface with
larger micrometer-sized holes
and an underlying diffusion gap.
When a drop of fluid containing
microbeads is placed on the top
surface, the beads become trapped
in the micrometer-sized holes while
the fluid is free to flow through the
diffusion layer and out of the array.
This structure has the advantage of
allowing beads of different sizes to
be trapped and permanently fixed
in different parts of the device as
the fluid evaporates (see image).
We studied how a droplet of
liquid evaporates and how this
affects the flow field, says Wong.
Based on simulations and experi-
ments, we were able to optimize our
microtrap architecture for efficient
size-based sorting of a range of
different bead sizes.
The researchers expect their
fabrication method to alleviate
ease-of-use issues associated with
current bead sorting assays and
also to significantly speed-up
throughput by allowing multiple
molecular targets to be detected in
one device. The additional dimen-
sion of bead size would directly
increase the number of analytes
that can be detected, says Wong.
They could increase from two, for
a conventional two-color system,
to six for a system with three bead
sizes in different trap regions.
1. Wong, C. C., Liu, Y., Wang, K. Y. &
Rahman, A. R. A. Size based sorting
and patterning of microbeads by
evaporation driven flow in a 3D
micro-traps array. Lab on a Chip 13,
36633667 (2013).
CHEMISTRY &
MATERIALS
Electronic transistors, which act as
miniature switches for controlling
the flow of electrical current,
underpin modern-day microelec-
tronics and computers. State-of-the-
art microprocessor chips contain
several billion transistors that
switch signals flowing in electrical
wires and interconnects (see image).
With increasing data-processing
speeds and shrinking chip sizes,
however, wires and interconnects
waste considerable energy as heat.
One alternative is to replace
electrical interconnects with
energy-efficient optical intercon-
nects that carry data using light
signals. However, a practical ana-
logue of the transistor for optical
interconnects does not yet exist.
Hence, Vivek Krishnamurthy from
the A*STAR Data Storage Institute
and co-workers in Singapore and
the United States are developing
a practical photonic transistor
for optical interconnects that can
control light signals in a similar
manner to electronic transistors.
The researchers latest photonic
transistor design is based on
prevalent semiconductor technology
and offers attractive attributes of
high switching gain, low switching
power and high operating speed
1
.
Importantly, the research teams
design enables a switching gain of
greater or equal to 2, which means
the output signal is more than
double the strength of the input
signal. Hence, the transistor can be
cascaded: the output signal from
one photonic transistor is suffi-
ciently strong so that it can be split
to feed several others. Known as
fan-out, this functionality means
the design can become a building
block to be scaled up to form larger
circuits with many such switching
elements connected together for
all-optical processing on an optical
interconnect platform for data- and
telecommunications. Furthermore,
Krishnamurthy says that the
design consumes 1020 times
less power than the conventional
all-optical switching technologies
and can operate at very fast speeds.
The teams design consists of
a circuit of coupled silicon wave-
guides that guide infrared light with
a wavelength of 1.5 micrometers.
Some of the waveguides feature an
optically active material, such as an
indium gallium arsenide semicon-
ductor, that can amplify or absorb
signal light depending on whether
or not it is optically excited. During
operation, the intensity of a short-
wavelength routing beam is used to
control the strength of an output
beam by altering the amount of
absorption and gain in the circuit.
The researchers are now
working to experimentally realize
their optical transistor. We are
realizing it on a silicon chip so that
it will be compatible with current
microelectronic industry standards
to enable commercial deploy-
ment, explains Krishnamurthy.
Once we experimentally
verify the prototype, we could
further integrate it into large-scale
optical switching systems for
optical interconnects.
Optoelectronics:
A transistor for light
A high-performance photonic transistor that switches light signals instead
of electronic signals could revolutionize optical signal processing
The realization of practical photonic transistors could revolutionize optical signal processing in the same way as the invention of
the electronic transistor in the 1950s did for microelectronics. Intels state-of-the-art coprocessor, the Xeon Phi (above), contains a
staggering five billion electronic transistors.

I
n
t
e
l

C
o
r
p
o
r
a
t
i
o
n
1. Krishnamurthy. V., Chen. Y. &
Ho S.-T. Photonic transistor
design principles for switching
gain >=2.Journal of Lightwave
Technology 31, 20862098 (2013).
CHEMISTRY &
MATERIALS
The semiconductor silicon lies at
the heart of the current revolution
in electronics and computing. In
particular, it can produce compact
integrated circuits when processed
by modern techniques capable of
fabricating structures just a few
nanometers in size.
Now, Man-Fai Ng and
Teck Leong Tan at the A*STAR
Institute of High Performance
Computing in Singapore have
shown that mixing silicon with
similar materials can open
the door to the fabrication of
nanoscale devices with a diverse
array of properties that have a
wider range of applications
1
.
Ng and Tan used state-of-
the-art computer simulations to
assess the structural stability and
electronic properties of silicon-
based nanowires. As their name
suggests, nanowires are just a few
nanometers wide but can be up to
a millimeter long. They exhibit
unusual electronic properties
because their small width confines
the motion of electrons across
the wire.
The properties of silicon
nanowires are well established,
but there is considerable scope
to expand their applicability.
Scientists anticipate they could
realize a more diverse range
of characteristics by partially
replacing silicon with other
elements that are in the same
column as silicon in the periodic
table. There are many potential
materials including carbon,
germanium and tin each of
which can be combined with
silicon in any ratio to form
an alloy.
Consequently, the total
number of possible alloys is
immense. The researchers thus
undertook a comprehensive search
of all these silicon-based alloys to
determine which are atomically
stable and which have the best
properties for nanowire devices.
Ng and Tan employed three
mathematical techniques (namely,
density functional theory, the
cluster expansion method and the
Monte Carlo method) to simulate
different atomic arrangements
in nanowires.
Instead of evaluating all
possible alloy structures, our
multiscaled simulation approach
enabled rapid large-scale com-
parison of different combinations
of alloy structures and selected the
thermodynamically stable ones,
explained Ng.
The most stable germanium
silicon and tinsilicon nanowires
were found to be those in which
the silicon atoms are concentrated
around the edge of the wire and
the other atomic species are at the
core. Conversely, an optimum
carbonsilicon nanowire exhibited
an ordered arrangement of the
atomic species (see image).
Once they had identified the
optimum atomic arrangement,
Ng and Tan calculated the energy
bandgap a critical parameter
for determining the electronic
properties of semiconductors.
Next, we plan to improve the
bandgap prediction for silicon-
based nanowires and develop
our approach to address more
complicated nanosystems for
energy applications, says Ng.
Nanotechnology:
Finding the right mix
Computer simulations indicate that mixing silicon with other materials
improves the diversity of nanoscale electronic devices
Cross-sectional view of stable nanowires made from carbonsilicon (left), germanium
silicon (center) and tinsilicon (right), as predicted by calculations. The silicon atoms (yel-
low) are found at the edge of the nanowire when alloyed with tin (gray) and germanium
(green). In contrast, in carbonsilicon nanowires (where carbon is indicated in black), they
have an ordered arrangement.

2
0
1
4

A
*
S
T
A
R

I
n
s
t
i
t
u
t
e

o
f

H
i
g
h

P
e
r
f
o
r
m
a
n
c
e

C
o
m
p
u
t
i
n
g
1. Ng, M.-F. & Tan, T. L. Unveiling stable
group IV alloy nanowires via a
comprehensive search and their
electronic band characteristics.
Nano Letters 13, 49514956 (2013).
Next, we plan to improve
the bandgap prediction
for silicon-based
nanowires and develop
our approach to address
more complicated
nanosystems for
energy applications.
CHEMISTRY &
MATERIALS
Metamaterials are engineered to
interact with light and sound waves
in ways that natural materials
cannot. They thus have the poten-
tial to be used in exciting new
applications, such as invisibility
cloaks, high-resolution lenses,
efficient and compact antennas,
and highly sensitive sensors.
While the theory of this
interaction is relatively well under-
stood, it has been challenging to
fabricate metamaterials that are
large enough to be practical. Now,
Yi Zhou and colleagues at the
A*STAR Data Storage Institute
in Singapore have demonstrated
a promising new fabrication
technique that can produce
large areas of an important
class of metamaterial, known as
fishnet metamaterials
1
.
Most optical metamaterials
consist of tiny repeated metallic
structures. When light of a
particular frequency falls on
them, it establishes oscillating
fields inside each structure.
These fields can resonate with
each other and thereby produce
desirable collective behavior.
Fishnet metamaterials usually have
several vertically stacked repeat
units spread out over much larger
lateral dimensions. Because they
are structured both vertically and
laterally, they are called three-
dimensional materials.
Fishnet metamaterials are
usually made in one of two ways.
They can be fabricated by care-
fully patterning individual films
and then stacking these films on
top of each other. However, this
multilayer process is difficult, as
it requires careful alignment of
the films.
The second approach is to
pattern a sacrificial substrate and
then deposit repeated layers onto
it. This pattern-first process
suffers from its own difficulties,
the most important of which is
that the total thickness of the
final fishnet material is typically
limited to tens of nanometers or
less. This restricts the kind of
resonances that can be achieved
and, in turn, the functionality of
the final film.
Zhou and colleagues were able
to increase the total thickness
of pattern-first fishnet films
to around 300 nanometers,
allowing five bilayers of film to
be deposited and resulting in a
strong characteristic resonance
and pronounced metamaterial
behavior. To achieve this, they
adopted a technique called trilayer
lift-off, which is commonly used
in industry but seldom applied in
research laboratories. It involves
patterning a sacrificial layer of a
photoresist resting on a layer of
silicon dioxide under which lies a
second photoresist layer.
By alternating the patterning
and etching steps, the A*STAR
team could achieve a film thickness
greatly exceeding the size of the
lateral patterns etched into the film.
This technique will help researchers
design large-area three-dimensional
nanodevices more easily, says
Zhou, and help bring the science of
metamaterials to reality.
Metamaterials:
Making it big
The use of a fabrication technique borrowed from the semiconductor
industry brings metamaterial applications a step closer to reality
Metamaterials can enable novel applications, like superlenses, which overcome the
diffraction limits of normal optics.

L
e
a
f
e
n
L
i
n
/
i
S
t
o
c
k
/
T
h
i
n
k
s
t
o
c
k
1. Zhou, Y., Chen, X. Y., Fu, Y H.,
Vienne, G., Kuznetsov, A. I., &
Lukyanchuk, B. Fabrication of
large-area 3D optical fishnet
metamaterial by laser interference
lithography. Applied Physics Letters
103, 123116 (2013).
This technique will
help researchers design
large-area three-
dimensional nanodevices
more easily, and help
bring the science of
metamaterials to reality.
CHEMISTRY &
MATERIALS
Magnetic refrigeration is attracting
attention as an efficient way to chill
sensitive scientific instruments.
This refrigeration method exploits
the magnetocaloric effect, in which
an external magnetic field controls
the temperature of a magnetic mate-
rial. Effective magnetic refrigerants
are often difficult to prepare, but
now Andy Hor of the A*STAR
Institute of Materials Research
and Engineering and the National
University of Singapore and his
colleagues have created a powerful
magnetic refrigerant that is easy to
make in the lab
1
.
Compounds with a large
magnetocaloric effect typically
contain atoms with many unpaired
electrons, each of which generates
its own tiny magnetic moment.
During magnetic refrigeration, an
external magnetic field forces these
atomic magnetic moments to line
up in the same direction. As the
magnetism of the atoms becomes
more ordered (which reduces the
entropy of the system), the mate-
rials temperature rises.
Once the heat has been removed
by a flowing liquid or gas, the
external magnetic field is reduced.
This allows the atomic magnetic
moments to become disordered
again, cooling the material so that
it can be used to draw heat from
an instrument, before repeating
the cycle.
Magnetic refrigerants commonly
use the gadolinium(III) ion (Gd
3+
),
because it has seven unpaired elec-
trons. Most gadolinium complexes
are made under harsh conditions
or take a very long time to form,
which limits their wider application.
In contrast, the magnetic refrigerant
developed by Hor and colleagues is
remarkably easy to make.
The researchers simply mixed
gadolinium acetate, nickel acetate
and an organic molecule called
2-(hydroxymethyl)pyridine
in an organic solvent at room
temperature. After 12 hours, these
chemicals had assembled themselves
into an aggregate containing a
cube-like structure of atoms at its
heart (see image).
The team measured how an
external magnetic field affected
this cubane material as the
temperature dropped. Below about
50 K, they found that the materials
magnetization increased sharply,
suggesting that it could be an
effective magnetic refrigerant below
this temperature.
The scientists then tested the
effects of varying the external
magnetic field at very low
temperatures. They found that at
4.5 K, a large external field caused
an entropy change that was close
to the theoretical maximum for
the system and larger than most
other magnetic refrigerants under
similar conditions.
According to the team, the
magnetocaloric effect of magnetic
refrigerants has typically been
enhanced by creating ever-larger
clusters of metal atoms. In
contrast, their cubane shows
that much simpler aggregates,
prepared under straightforward
conditions, are promising as
magnetic refrigerants.
The magnetic refrigerant contains a cubic structure made of two gadolinium ions (pink),
two nickel ions (green) and four oxygen atoms (red), surrounded by 2-(hydroxymethyl)
pyridine molecules.
Materials:
Cubic cluster chills out
1. Wang, P., Shannigrahi, S.,
Yakovlev, N. L., and Hor, T. S. A.
Facile self-assembly of
intermetallic [Ni
2
Gd
2
] cubane
aggregate for magnetic
refrigeration. Chemistry An Asian
Journal 8, 29432946 (2013).
Gadolinium-based material that can be cooled by varying a magnetic eld
may be useful for cooling low-temperature sensors

2
0
1
4

A
*
S
T
A
R

I
n
s
t
i
t
u
t
e

o
f

M
a
t
e
r
i
a
l
s

R
e
s
e
a
r
c
h

a
n
d

E
n
g
i
n
e
e
r
i
n
g
CHEMISTRY &
MATERIALS
Plasmonic devices such as
superlenses, hyperlenses and
plasmonic waveguides have
exciting potential for research and
commercial applications because
they permit optical lithography,
imaging and waveguiding to be
performed at resolutions below the
diffraction limit of light. These
devices often require low-loss
ultrathin metal films, which are
difficult to fabricate using current
deposition techniques. Researchers
have investigated processes such as
seed layer deposition and thermal
annealing to reduce the surface
roughness and grain-boundary
density of these films. To date,
however, these processes have not
been hugely successful.
Now, Ee Jin Teo and colleagues
at the A*STAR Institute of Mate-
rials Research and Engineering,
Singapore, the University of Hyogo,
Japan, and the National University
of Singapore have used gas cluster
ion beam (GCIB) processing to
smooth ultrathin metal films and
thereby enhance their properties
1
.
A GCIB consists of thousands
of gas molecules that are weakly
bound by van der Waals forces.
Such a beam is able to smooth out
surface irregularities and reduce
film thickness with nanometer
precision. This processing signifi-
cantly enhances surface plasmon
resonance and propagation, and
enables the fabrication of ultrathin
films with extremely low electrical
resistivity and optical loss.
Unlike monomer ion
beams used in conventional
ion-beam milling and plasma
etching, a cluster of nitrogen
gas molecules with an energy of
20 kiloelectron volts impinging
on a silver film can deliver a high
energy density to a relatively
small volume: yet the cluster
penetrates to a depth of only a
few nanometers. The impact of
the beam on the film causes silver
atoms in surface peaks to scatter
sideways toward valleys, voids
and grain boundaries. As well as
producing a smoother surface, this
processing triples the grain width
through the redeposition of atoms
at grain boundaries.
The teams GCIB treatment
resulted in up to a four-fold
improvement in the electrical and
optical properties of films of a
thickness of 12 nanometers. The
unique characteristics of GCIB
irradiation meant that in a single
irradiation step we could reduce
scattering losses due to surface
roughness, grain boundaries and
voids, notes Teo.
The research team also used the
technique to smooth the top surface
and sidewalls of lithographically
patterned silver-stripe waveguides,
increasing the propagation
lengths of surface plasmons in
these waveguides.
In the future, we intend to
use this technique to improve the
color purity of plasmonic color
filters or reflectors, and also to
increase the patterned area of
superlens nanolithography, says
Teo. Such developments will bring
plasmonic research a step closer
to commercialization.
Plasmonics:
Minimizing loss by thinning
and smoothing
Gas cluster ion beam smoothing produces ultrathin silver lms and
lithographically patterned structures to enhance plasmonic performance
The smoothing effect of a gas cluster ion beam (purple) on a rough surface (gray).

2
0
1
4

A
*
S
T
A
R

I
n
s
t
i
t
u
t
e

o
f

M
a
t
e
r
i
a
l
s

R
e
s
e
a
r
c
h

a
n
d

E
n
g
i
n
e
e
r
i
n
g
1. Teo, E. J., Toyoda, N., Yang, C.,
Wang, B., Zhang, N. et al.
Sub-30 nm thick plasmonic films
and structures with ultralow loss.
Nanoscale 6, 32433249 (2014).
CHEMISTRY &
MATERIALS
Natural surfaces that repel water,
such as lotus leaves or butterfly
wings, often have a rough,
microscale texture that traps air
beneath the liquid droplet. By
mimicking these biological struc-
tures, researchers have developed
superhydrophobic coatings that
are highly resistant to wetting. One
trick unknown to nature, however,
is the ability to repel hydrocarbon-
based oils that have much lower
surface tension than water and tend
to spread out rather than bead up.
Jia Min Chin and co-workers
from the A*STAR Institute
of Materials Research and
Engineering and A*STAR
Institute of Bioengineering and
Nanotechnology in Singapore
have now discovered a simple
procedure to synthesize omni-
phobic interfaces that repel both
oil and water using intricate,
mushroom-shaped, metalorganic
crystal frameworks
1
.
Recent efforts toward omni-
phobic surfaces have focused on
producing reentrant microscale
textures, which have curved
shapes that inherently retain air
pockets. These structures prevent
oil from wetting the surface and
stabilize the beaded droplet state.
Currently, complicated and labor-
intensive lithographic fabrication
techniques are needed to generate
such textures.
Chin and co-workers inves-
tigated a bottom-up strategy
to synthesize omniphobic films
using metalorganic frameworks
(MOFs) compounds
that connect metal ions into
multidimensional structures using
hydrocarbon-based linkages.
Previous studies have shown that
an aluminum-containing MOF,
known as NH
2
-MIL-53(Al), can
controllably form micro- and
nanoscale rods and needles. The
team suspected that suitable
synthetic conditions could yield
spontaneous needle growth
upward from a substrate, forming
a micro-rough surface with
numerous trapped air pockets.
To achieve this, the researchers
mixed their MOF precursor with
an aluminum oxide membrane
and applied hydrothermal
high temperaturehigh pressure
aqueous reaction conditions. This
resulted in perpendicularly aligned
needles on both sides of the
membrane. Next, the team faced
the challenge of transforming
the needles into curved textures
suitable for repelling oil. After
many attempts, they spotted an
important clue the modified
membranes floated on top of
aqueous surfaces due to their
superhydrophobic nature.
Chin and her team exploited
this floating effect by suspending
the microneedle-covered mem-
brane in an aqueous solution of
the MOF precursor. Additional
MOF growth occurred only on
the wetted tips of the needles,
expanding the crystalline stems
into mushroom-like caps
(see image). By controlling the
reaction time to generate a targeted
cap size, the researchers omni-
phobic surface successfully repelled
long-chain hydrocarbon oils.
Chin notes that this benchtop,
chemical process produces results
previously limited to facilities
with expensive, high-tech equip-
ment. Our aim was to develop
simple techniques for fabricating
interesting structures which are
accessible to scientists around the
world, she says.
Surface engineering:
Metalorganic
micromushrooms repel all
A clever chemical transformation yields surface-bound microstructures that
efciently drive away oil- and water-based contaminants
Scanning electron microscopy images (left) reveal how the transformation of a
metalorganic microneedle surface (top) into a micromushroom interface (bottom)
causes oil droplets (right) to bead up instead of spreading out.
M
o
d
i
e
d

f
r
o
m

R
e
f
.
1

2
0
1
3

A
m
e
r
i
c
a
n

C
h
e
m
i
c
a
l

S
o
c
i
e
t
y

1. Tan, T. T. Y., Reithofer, M. R., Chen, E. Y.,
Menon, A. G., Hor, T. S. A., Xu, J. &
Chin, J. M. Tuning omniphobicity via
morphological control of
metalorganic framework
functionalized surfaces. Journal of
the American Chemical Society 135,
1627216275 (2013).
CHEMISTRY &
MATERIALS
The rise of drug-resistant
microbes is a major challenge
facing medicine. The World
Health Organizations 2014
report on global surveillance of
antimicrobial resistance warns
of the very real possibility of the
twenty-first century becoming
a post-antibiotic era in which
common infections and minor
injuries can kill. In the face of
this threat, researchers worldwide
are exploring approaches to find
new compounds that combine
selective antimicrobial efficacy
with low toxicity toward mam-
malian cells.
Yi Yan Yang at the A*STAR
and co-workers have now created
a range of large polycarbonate
molecules that are potent anti-
microbials and are tolerated well
by rat red blood cells, suggesting
that they could prove similarly
effective in humans
1
. Crucially, by
subtly varying the composition of
the polycarbonate molecules, the
researchers could fine-tune the
selectivity and activity of these
candidate drugs.
Antimicrobial polycarbonates
are long-chain polymers made by
linking small monomer molecules.
Each monomer contains two com-
ponents: one that is hydrophobic
and physically inserts into the cell
membrane of bacteria and fungi;
and one that carries a positively
charged group that is attracted to
the negative charge on the surface
of microbial cells.
By carefully tinkering with
the hydrophobic and hydrophilic
balance between the components
of these new monomers, the
researchers were able to create
polymers that adhere to microbial
cells and disrupt their cell mem-
branes, thereby killing the cells
(see image).
The polycarbonates developed
by the researchers have proven
highly effective against a variety
of clinically isolated multidrug-
resistant bacteria and fungi. A
further benefit is that the mol-
ecules are biodegradable, which
means that, when used in clinical
situations, they should take effect
and then degrade naturally.
This attribute provides a crucial
advantage over other synthetic
alternatives that persist and cause
undesirable side effects.
Using scanning electron
microscopy, the researchers
showed that the molecules work
by breaking open the microbial
cell membrane a mode of
action they believe reduces the
likelihood of microbes becoming
resistant to the polycarbonates.
The ability to explore
different compositions
within the monomers may
allow further enhancements,
according to the researchers.
By carefully controlling the
structure and the ratio of the two
components, we can enhance
dramatically the selectivity of
the polymers toward a broad
range of pathogenic microbes,
says Yang. The researchers will
now study the in vivo efficacy
of the optimal polymer using
intravenous injection into mice
infected with methicillin-resistant
Staphylococcus aureus (MRSA)
bacteria that have developed
resistance to a broad class
of antibiotics.
Antimicrobials:
Polycarbonates to tackle
multidrug resistance
Polymers that can be ne-tuned for optimal efect could help ght
multidrug-resistant infections
1 m
Scanning electron microscopy images of Escherichia coli before (top) and after (bottom)
two-hour treatment with a polymer. The treated E. coli cells show distorted and cor-
rugated surfaces compared to the intact control cells.
R
e
p
r
o
d
u
c
e
d
,
w
i
t
h

p
e
r
m
i
s
s
i
o
n
,
f
r
o
m

R
e
f
.
1

2
0
1
3

A
m
e
r
i
c
a
n

C
h
e
m
i
c
a
l

S
o
c
i
e
t
y
1. Chin, W., Yang, C, Ng, V. W. L.,
Huang, Y., Cheng, J. et al.
Biodegradable broad-spectrum
antimicrobial polycarbonates:
Investigating the role of chemical
structure on activity and
selectivity. Macromolecules
46, 87978807 (2013).
By carefully controlling
the structure and
the ratio of the two
components, we can
enhance dramatically
the selectivity of the
polymers toward a broad
range of pathogenic
microbes.
CHEMISTRY &
MATERIALS
Dye-sensitized solar cells (DSSCs)
rely on dyes that absorb light to
mobilize a current of electrons
and are a promising source of
clean energy. Jishan Wu at the
A*STAR Institute of Materials
Research and Engineering and
colleagues in Singapore have now
developed zinc porphyrin dyes
that harvest light in both the
visible and near-infrared parts
of the spectrum
1
. Their research
suggests that chemical modifica-
tion of these dyes could enhance
the energy output of DSSCs.
DSSCs are easier and cheaper
to manufacture than conventional
silicon solar cells, but they
currently have a lower efficiency.
Ruthenium-based dyes have been
traditionally used in DSSCs, but
in 2011 researchers developed a
more efficient dye based on a zinc
atom surrounded by a ring-shaped
molecule called a porphyrin. Solar
cells using this new dye, called
YD2-o-C8, convert visible light
into electricity with an efficiency
of up to 12.3 per cent. Wus team
aimed to improve that efficiency
by developing a zinc porphyrin
dye that can also absorb
infrared light.
The most successful dyes
developed by Wus team,
WW-5 and WW-6, unite a zinc
porphyrin core with a system of
fused carbon rings bridged by
a nitrogen atom, known as an
N-annulated perylene group.
Solar cells containing these dyes
absorbed more infrared light than
YD2-o-C8 and had efficiencies of
up to 10.5 per cent, matching the
performance of an YD2-o-C8 cell
under the same testing conditions
(see image).
Theoretical calculations
indicate that connecting the
porphyrin and perylene sections
of these dyes by a carboncarbon
triple bond, which acts as an
electron-rich linker, improved the
f low of electrons between them.
This bond also reduced the light
energy needed to excite electrons
in the molecule, boosting the dyes
ability to harvest infrared light.
Adding bulky chemical groups
to the dyes also improved their
solubility and prevented them
from aggregating something
that tends to reduce the efficiency
of DSSCs.
However, both WW-5 and
WW-6 are slightly less efficient
than YD2-o-C8 at converting
visible light into electricity, and
they also produce a lower voltage.
We are now trying to solve this
problem through modifications
based on the chemical structure of
WW-5 and WW-6, says Wu.
Comparing the results from
more peryleneporphyrin dyes
should indicate ways to overcome
these hurdles, and may even
extend light absorption further
into the infrared. The top
priority is to improve the power
conversion efficiency, says Wu.
Our target is to push the effi-
ciency to more than 13 per cent in
the near future.
Energy:
Dyes help harvest light
Dye-sensitized solar cell absorbs a broad range of visible and
infrared wavelengths
Zinc porphyrin dyes were used to create
solar cells that can absorb both visible
and near-infrared light.

2
0
1
4

A
*
S
T
A
R

I
n
s
t
i
t
u
t
e

o
f

M
a
t
e
r
i
a
l
s

R
e
s
e
a
r
c
h

a
n
d

E
n
g
i
n
e
e
r
i
n
g
1. Luo, J., Xu, M., Li, R., Huang, K.-W.,
Jiang, C. et al. N-annulated perylene
as an efficient electron donor for
porphyrin-based dyes: Enhanced
light-harvesting ability and high-
efficiency Co(II/III)-based dye-
sensitized solar cells. Journal of the
American Chemical Society 136,
265272 (2014).
CHEMISTRY &
MATERIALS
Increasing the cost-effectiveness
of photovoltaic devices is
critical to making these renewable
energy sources competitive with
traditional fossil fuels. One pos-
sibility is to use hybrid solar cells
that combine silicon nanowires
with low-cost, photoresponsive
polymers. The high surface area
and confined nature of nanowires
allows them to trap significant
amounts of light for solar cell
operations. Unfortunately, these
thin, needle-like structures are very
fragile and tend to stick together
when the wires become too long.
Now, findings by Xincai Wang
from the A*STAR Singapore
Institute of Manufacturing
Technology and co-workers
from Nanyang Technological
University could turn the tables
on silicon nanowires by improving
the manufacturing of silicon
nanoholes narrow cavities
carved into silicon wafers that
have enhanced mechanical and
light-harvesting capabilities
1
.
Nanoholes are particularly
effective at capturing light because
photons can ricochet many times
inside these openings until absorp-
tion occurs. Yet a practical under-
standing of how to fabricate these
tiny structures is still lacking. One
significant problem, notes Wang,
is control of the initial stages of
nanohole formation a crucial
period that can often induce
defects into the solar cell.
Instead of traditional
time-consuming lithography,
the researchers identified a
rapid, maskless approach to
producing nanoholes using silver
nanoparticles. First, they deposited
a nanometer-thin layer of silver
onto a silicon wafer, which they
toughened by annealing it using
a rapid-burst ultraviolet laser.
Careful optimization of this
procedure yielded regular arrays
of silver nanospheres on top of the
silicon surface, with sphere size
and distribution controlled by the
laser annealing conditions.
Next, the nanospheresilicon
complex was immersed into a
solution of hydrogen peroxide and
hydrofluoric acid a mixture
that eats away at silicon atoms
directly underneath the catalytic
silver nanospheres. Subsequent
removal of the silver particles with
acid produced the final, nanohole-
infused silicon surface (see image).
The team analyzed the solar
cell activity of their nanohole
interfaces by coating them
with a semiconducting polymer
and metal electrodes. Their
experiments revealed a remark-
able dependence on nanohole
depth: cavities deeper than one
micrometer showed sharp drops in
power conversion efficiency from
a maximum of 8.3 per cent due
to light scattering off of rougher
surfaces and higher series
resistance effects.
Our simple process for
making hybrid silicon nanohole
devices can successfully reduce
the fabrication costs which
impede the solar cell industry,
says Wang. In addition, this
approach can be easily transferred
to silicon thin films to develop
thin-film siliconpolymer
hybrid solar cells with even
higher efficiency.
Solar cells:
Powered by nanoholes
A simple and inexpensive fabrication procedure boosts the light-capturing
capabilities of tiny holes carved into silicon wafers
200 nm
A straightforward procedure that transforms silver nanospheres (top) into silicon nanoholes (bottom) can overcome the shortcomings
of nanowire-based solar cells.
R
e
p
r
o
d
u
c
e
d
,
w
i
t
h

p
e
r
m
i
s
s
i
o
n
,
f
r
o
m

R
e
f
.1

2
0
1
4

A
m
e
r
i
c
a
n

I
n
s
t
i
t
u
t
e

o
f

P
h
y
s
i
c
s
1. Hong, L., Wang, X., Zheng, H., He, L.,
Wang, H., Yu, H. & Rusli, E. High
efficiency silicon nanohole/organic
heterojunction hybrid solar cell.
Applied Physics Letters 104,
053104 (2014).
CHEMISTRY &
MATERIALS
Plants and herbs containing
phthalide molecules organic
compounds that fuse aromatic
benzene rings with cyclic,
oxygen-bearing hydrocarbons
known as lactones have
been used since ancient times
as natural medicines to treat
diseases such as cardiovascular
ailments. Chemists have recently
discovered that phthalides can
act as valuable scaffolds in the
construction of pharmaceuticals
with complex structures and more
potent capabilities. Unfortunately,
phthalides are normally produced
through multistep synthetic
pathways that can significantly
increase manufacturing costs.
Jayasree Seayad and co-workers
from the A*STAR Institute
of Chemical and Engineering
Sciences in Singapore have now
developed a novel catalytic process
that can generate numerous
phthalide compounds in a
single step from readily available
feedstocks a green chemistry
approach that simultaneously
boosts synthetic efficiency and
reduces production of environ-
mentally hazardous waste
1
.
One way that chemists are
trying to improve the effective-
ness of their reactions is with
carbonhydrogen (CH) activa-
tion. Typically, CH bonds are
inert and require harsh conditions
to cleave. By contrast, CH
activation reactions can transform
these bonds into new functional
groups under very mild conditions
using catalytically active transition
metals such as rhodium complexes.
Seayad and her team antici-
pated that benzaldehyde a
common benzene derivative
bearing a simple carbonyl side
chain known as an aldehyde
could serve as ideal reagents for
phthalide synthesis because of
their structural similarity. To
achieve this goal, however, the
researchers had to find a way to
selectively activate the benzalde-
hyde CH bond located beside
the aldehyde chain. This is a
challenge, notes Seayad, because
aldehyde groups are poor at
directing the transition metal
catalyst to this site.
To overcome this obstacle, the
team turned to a combination
of rhodium and organic amine
catalysts (see image). The amine
catalyst first activates benzaldehyde
by replacing its oxygen atom with
nitrogen, forming a functional
group known as an imine. The
imine bonds rapidly to the rho-
dium catalyst and positions it for
favorable CH activation. Then,
a second benzaldehyde molecule
couples with the activated complex
to generate the desired phthalide
in high yields. Performing this
reaction under oxidative condi-
tions ensures easy regeneration of
the rhodiumamine system for
numerous catalytic cycles.
The researchers found that this
strategy enabled both identical
and modified benzaldehydes to
couple into phthalides, opening
the way for chemists to produce
a diverse range of these bioactive
molecules with ease. This
discovery will lead to simplified
synthesis of phthalide intermedi-
ates from readily available alde-
hydes while avoiding multistep
pathways, says Seayad.
Catalysis:
Modern routes to ancient
remedies
An innovative catalytic system makes the synthesis of sought-after
bioactive molecules easier and more efcient than ever
O
O
O
H
+
R CHO
H
R
Oxidative
conditions
Phthalide
Benzaldehyde
molecules
Dual catalysis
Rhodium complex
+ amine
A combination of transition metal and organic catalysts can help benzaldehyde
molecules to combine into bioactive phthalides in a single step.

2
0
1
4

A
*
S
T
A
R

I
n
s
t
i
t
u
t
e

o
f

C
h
e
m
i
c
a
l

a
n
d

E
n
g
i
n
e
e
r
i
n
g

S
c
i
e
n
c
e
s
1. Tan, P. W., Juwaini, N. A. B. &
Seayad, J. Rhodium(III)-amine
dual catalysis for the oxidative
coupling of aldehydes by directed
CH activation: Synthesis of
phthalides. Organic Letters 15,
51665169 (2013).
This discovery will lead
to simplied synthesis of
phthalide intermediates
from readily available
aldehydes while avoiding
multistep pathways.
CHEMISTRY &
MATERIALS
Thanks to the positively and
negatively charged units in
their monomers, zwitterionic
polymers have a high affinity
for water a property known
as hydrophilicity. This property
helps prevent fouling, namely
the build-up of contaminants.
Current zwitterionic polymers are
not effective in water as they use
monomers such as commercially
available acrylamide and meth-
acrylates that tend to decompose
and lose their electrostatic
characteristics when wet.
To solve this issue, a team
led by Vivek Vasantha from the
A*STAR Institute of Chemical
and Engineering Sciences in
Singapore has now developed
zwitterionic polymers based
on water-stable monomers that
incorporate nitrogen-containing
derivatives known as imidazoles
1
.
The team introduced the zwit-
terions to readily accessible,
hydrophobic polystyrene to
boost its hydrophilicity in water
by forming a hydration layer
through electrostatic interactions
and hydrogen bonding.
To synthesize the monomers,
Vasanthas team reacted styrene
precursors with positively
charged imidazoles before
attaching the negatively charged
sulfonate functional groups.
The monomers produced
polymers with intact zwitterionic
properties, meaning that they
retained their positive and
negative charges.
These new imidazole-based
polymers exhibited some novel
solubility characteristics: unlike
their conventional water-soluble
counterparts, they swelled in
water and dissolved only in
highly concentrated brine.
These differences stem from
dipoledipole interactions and
the more hydrophobic nature of
the new polymers compared to
acrylamide and methacrylate.
With high tolerances to salt,
pH and temperature, these
polymers became increasingly
viscous when subjected to higher
shear forces in brine. This
characteristic similar to silly
putty, which is malleable in ones
hands but is unchanged when
hit with a hammer makes the
polymers attractive for enhanced
oil recovery and marine anti-
fouling coatings.
Another advantage of the
new polymers is their reversible
phase change: between 5 C and
95 C, the polymers formed gels
that become clear f luids when
heated above the so-called critical
temperature in brine and that
revert to their stable cloudy state
on cooling.
This phase transition
results from the disruption of
the equilibrium between salt,
water and zwitterionic species,
says Vasantha. The polymer
chains expand on heating and
collapse below the critical tem-
perature. The researchers could
control the critical temperature
by simply varying either the brine
or polymer concentration. For
example, the transition occurred
at 20 C at a low polymer concen-
tration but at 40 C at a higher
polymer concentration.
We are currently designing
new zwitterionic polymers
and copolymers with salt- and
heat-responsive behavior for a
wide range of applications, such
as enhanced oil recovery, low-
temperature protein separation
and antifouling, says Vasantha.
Thermoresponsive polymers:
Positive progress on
antifouling
Heat-responsive polymers that do not breakdown in water may lead to new
antifouling coatings and enhanced oil recovery
Their high tolerance to salt, pH and temperature cause zwitterionic polymers to become
viscous when subjected to high shear forces in brine, making them useful for marine
antifouling applications.

A
l
e
x
c
r
a
b
/
i
S
t
o
c
k
/
T
h
i
n
k
s
t
o
c
k
1. Vasantha, V. A., Jana, S.,
Parthiban, A. & Vancso, J. G.
Water swelling, brine soluble
imidazole-based zwitterionic
polymers synthesis and study
of reversible UCST behavior and
gelsol transitions. Chemical
Communications 50, 4648 (2014).
CHEMISTRY &
MATERIALS
Devices that manipulate
very small volumes of f luids
are applied in diverse fields,
including printer technology,
DNA processing and cooling
systems for electronics. For some
processes involving f luids, such
as mixing, it is useful to generate
oscillating f lows, but this can be
difficult for particularly viscous
f luids. Now, A*STAR researchers
have developed a microf luidic
oscillator that produces oscilla-
tions even in very viscous f luids
1
.
In miniaturized f luidic
devices, the viscous force of the
f luid dominates the f low, and
mixing becomes a challenging
task, says Huanming Xia from
the A*STAR Singapore Institute
of Manufacturing Technology
(SIMTech), who led the study
with co-workers at SIMTech and
the A*STAR Institute of High
Performance Computing. The
microf luidic oscillator is a part
of our continuous effort to solve
this problem.
Microf luidic valves and
pumps have diaphragms, which
are usually made from soft
materials, such as rubber, and
are operated via external forces.
Yet the tiny device, less than
4 millimeters in size, developed
by Xias team does not need
external control. Instead, when
the diaphragm is placed in a
f luid f low, it responds elasti-
cally by wiggling up and down
to make the device oscillate
automatically (see image). To
adapt the design for use with very
viscous f luids, the researchers
replaced the rubber diaphragm
with one made from copper and
beryllium foil.
While this device has
practical benefits, it also raises
theoretical implications about
the behavior of microf luidic
oscillators. The team found that
at low f luid pressures, the f low
across the diaphragm does not
oscillate. Then, above a particular
transition pressure, the f low
rate drops and oscillatory f low
occurs, increasing in frequency
as pressure increases. After
performing experimental and
theoretical tests for different
device shapes, f luid viscosities
and diaphragm thicknesses,
Xias team could expand cur-
rent theories.
Flow-induced vibrations
are usually related to flow
instabilities and analyzed using
a springmass model, explains
Xia. The transition from
laminar flow to oscillatory flow
in their new oscillator was coun-
terintuitive, because increased pres-
sure led to reduced flow rates. The
team recognized that this behavior
was similar to negative differential
resistance a well-established
concept that describes certain
electric circuits in which an
increased voltage leads to a
lower current.
Xias team is currently
developing a complete math-
ematical model of their device
using negative resistance and
other concepts borrowed from
electric circuit theory. This
should assist them to optimize
the device design for practical
applications; for example, the
enhanced mixing of viscous
f luids enabled by the device
can intensify and control
chemical reactions.
Microuidics:
Mixing through oscillations
A tiny device produces oscillatory ows that enhance the mixing of viscous
uids for chemical reactions
Diaphragm
Fluid
Oscillatory ow
Top (left) and side (right) views of the device. Fluid flow causes the diaphragm to
undergo elastic oscillations, producing an oscillatory flow at the outlet. R
e
p
r
o
d
u
c
e
d
,
w
i
t
h

p
e
r
m
i
s
s
i
o
n
,
f
r
o
m

R
e
f
.
1

2
0
1
3

A
I
P

P
u
b
l
i
s
h
i
n
g

L
L
C
1. Xia, H. M., Wang, Z. P., Nguyen, V. B.,
Ng, S. H., Wang, W. et al. Analyzing
the transition pressure and
viscosity limit of a hydroelastic
microfluidic oscillator. Applied
Physics Letters 104, 024101 (2014).
Flow-induced vibrations
are usually related to
ow instabilities and
analyzed using a
springmass model.
Research Highlights
GENETICS &
DISEASE
TECHNOLOGIES
GENETICS &
DISEASE
Found in the brain, the MeCP2 protein
plays a crucial role in the control of
hormones that regulate appetite and
metabolism. Deletion of MeCP2 may
increase cravings for fatty foods and lead
to obesity.

l
e
o
l
i
n
t
a
n
g
/
i
S
t
o
c
k
/
T
h
i
n
k
s
t
o
c
k
Obesity:
The power of proteins to battle
the bulge
The main cause of weight gain,
and ultimately obesity, is an energy
imbalance in the body triggered by
increased food intake, often coupled
with reduced energy expenditure.
Two hormones called leptin and
-MSH (-melanocyte-stimulating
hormone) regulate the so-called
energy homeostasis in the body
by influencing the brain to control
appetite, metabolism and behavior.
The hormone leptin acts on
at least two sets of neurons in the
brain, including a group called the
pro-opiomelanocortin (POMC)
neurons. Scientists know that
impaired POMC regulation leads to
leptin resistance and obesity, but are
unclear about the exact mechanism
by which this occurs.
Now, Weiping Han and col-
leagues at the A*STAR Singapore
Bioimaging Consortium and
A*STAR Institute of Molecular
and Cell Biology, together with
co-workers from Hungary and
the United States, have shown
that deleting a protein called
methyl-CpG-binding protein 2
(MeCP2) can promote weight gain
and obesity in mice
1
.
The team previously investigated
the onset of leptin resistance during a
high fat diet. We examined whether
epigenetic changes, such as DNA
methylation, could alter the gene
expression of the critical regulators
of energy homeostasis, notes Han.
POMC neurons are regulated by
MeCP2, so we chose to investigate
what would happen if this relation-
ship was disrupted.
The team bred genetically altered
mice that had MeCP2 deleted
specifically in the POMC neurons in
the hypothalamus region of the brain.
Their growth and feeding behavior
on a high fat diet were then compared
to their littermate control group with
fully functioning MeCP2. After
eight months, the MeCP2-knockout
mice showed higher levels of leptin
circulating in the blood, alongside
increased body weight and fat levels.
Under normal physiological
conditions, MeCP2, along with
other factors, allows increased
expression of POMC. This
occurs in response to cues from
hormones such as leptin, explains
Han. POMC is then processed to
generate -MSH, which is released
from POMC neurons to control
appetite and feeding behavior.
The researchers found that
without MeCP2 to increase POMC
expression in response to rising
leptin levels, not enough -MSH was
released to stem the appetite. This
led to all of the MeCP2-knockout
mice becoming overweight.
Knowing that increased DNA
methylation in POMC promoters
may lead to leptin resistance and
weight gain provides new options
for weight management, states
Han. Targeting epigenetic modi-
fication and regulation might be a
viable approach to treat obesity.
Hoping to better understand
the onset of obesity, the team is
now researching other proteins
that might affect the neurons and
hormonal pathways in the brain.
Improved understanding of the proteins that help the body to control
appetite may be useful for the treatment of obesity
1. Wang, X., Lacza, Z., Sun, Y. E. &
Han, W. Leptin resistance and
obesity in mice with deletion of
methyl-CpG-binding protein 2
(MeCP2) in hypothalamic pro-
opiomelanocortin (POMC) neurons.
Diabetologia 57, 236245 (2014).
GENETICS &
DISEASE
1. Skibinski, D. A. G., Hanson, B. J.,
Lin, Y., von Messling, V.,
Jegerlehner, A. et al. Enhanced
neutralizing antibody titers and
Th1 polarization from a novel
Escherichia coli derived pandemic
influenza vaccine. PLoS ONE 8,
e76571 (2013).
Inf luenza pandemics, such as the
2009 H1N1 swine f lu outbreak,
pose a serious risk to the global
population. Vaccination is one
route to protection but current
manufacturing methods for
vaccines limit the volume and
speed of production. Now, an
international team of researchers,
including A*STARs Program
in Translational Research on
Infectious Disease, Experimental
Therapeutics Centre and
Singapore Immunology Network,
has developed a more efficient
production process
1
.
In 2009, it took five months
for the vaccine against the H1N1
virus to become available, and
the number of doses made was
sufficient for only a fraction of the
global population. Both shortfalls
were down to the conventional
vaccine production process and
so the research team wanted to
develop a more efficient technique.
The licensed H1N1 f lu
vaccine is made by growing the
inf luenza virus in chicken eggs,
explains David Skibinski, the
studys lead author. In com-
parison, our vaccine is produced
by expressing the main surface
antigen of inf luenza in bacteria.
The researchers attached
the antigen part of a protein
from the surface of the H1N1
virus to harmless virus-like
particles and injected them into
mice (see image). Compared with
the mice that were inoculated
with the licensed vaccine, animals
inoculated with the new vaccine
produced a similar number of
effective antibodies against the
H1N1 virus, which demonstrated
that the new production method
provided immunity as effectively
as the existing vaccine.
There was, however, an
extra advantage: mice that were
inoculated with the new vaccine
produced more T cells than mice
that were inoculated with the
licensed vaccine. T cells play an
important role in protecting the
body against inf luenza, reduce the
severity of the disease, and pro-
vide protection against different
strains of the virus.
Skibinski says the new way of
producing the inf luenza vaccine
could improve the global response
to future pandemics by shortening
the time between the emergence
of a virus and production of a
functional vaccine, thus enabling
individuals to be vaccinated
before they become infected.
Greater cost-efficiency and
yields would also enable many
smaller or developing nations to
manufacture their own vaccine,
adds Skibinski.
Having shown the new
vaccines effectiveness in mice,
the research teams next step was
to see if the findings could be
transferred to humans. So far,
the results are positive, reports
John Connolly, the studys
senior author. A clinical trial
in humans under the direction
of the A*STAR D3 unit has
demonstrated that the vaccine is
safe, well tolerated and induces
antibody responses comparable
to those of approved seasonal
inf luenza vaccines.
Immunology:
Flu vaccine enables rapid
response
A more efcient production process for inuenza vaccines could help to
cope with future pandemics

2
0
1
4

A
*
S
T
A
R

I
n
s
t
i
t
u
t
e

o
f

M
o
l
e
c
u
l
a
r

a
n
d

C
e
l
l

B
i
o
l
o
g
y
100 nm
Electron micrograph of a virus-like particle-based influenza vaccine magnified 100,000 times.
TECHNOLOGIES
GENETICS &
DISEASE
The synthesis of proteins by
translation of RNA relies on a
protein complex called eIF4F.
As formation of this complex is
frequently unregulated in cancers,
the ability to restore its regulation
could pave the way to develop
new cancer therapies. A
Singapore-based team headed by
Christopher Brown and
David Lane from the A*STAR p53
Laboratory and Dilraj Lama at the
A*STAR Bioinformatics Institute
have used computer modeling to
design molecules that may be able
to do just that
1
.
Regulation of eIF4F involves
disrupting an interaction between
two proteins in the complex,
eIF4E and eIF4G. Brown and
colleagues set out to mimic this
disruption with artificial peptides,
or small protein fragments, that
bind to eIF4E in the same way as
eIF4G. To do this, they controlled
the peptide structures using an
approach called peptide stapling.
In the context of a whole
protein, peptides usually form
sections of localized structure,
but when these peptides are
removed from a protein they
become linear and highly f lexible,
explains Brown. A peptide
staple is a chemical modification
that can link two points of a
peptide together and reintroduce
the structure.
The research team designed
different stapled peptides and
ran computer simulations to
reveal their structures, both when
bound to eIF4E and when free in
solution. They also synthesized
the peptides and experimentally
characterized their structures
(see image).
Combining the information
obtained by both experimenta-
tion and simulation showed
where there was potential for
improvement. This meant
we could design new stapled
peptides, which could then be
further characterized through
experiments and also through
computer simulations, says
Brown. The new designs altered
individual amino acids. The aim
of these alterations was to either
optimize the peptide when bound
to eIF4E or, when in solution,
to make it look more like the
bound structure.
Using the optimization process,
the team produced two stapled
peptides that interacted strongly
with eIF4E. This required modi-
fication so that their structures
would stay the same in solution as
when bound to eIF4E.
We developed a process to
design highly potent inhibitors of
an interaction relevant to disease,
says Brown. Our next step is
to further understand how and
why certain stapled peptides are
biologically active whilst others
are not.
Although these stapled
peptides are not yet ready for use
in treatment, Brown hopes this
will soon change. In the future,
we envision developing the eIF4E
stapled interacting peptides into
biologically active molecules
and testing them in relevant
disease models.
The key steps in designing stapled peptides: modeling of peptides when unbound (left),
simulation of the peptide in complex with eIF4E to rationally design improved staples
(center), and experimental determination of the structure of the complex to validate the
method (right).
Biophysics:
Stapling together cancer therapy
1. Lama, D., Quah, S. T., Verma, C. S.,
Lakshminarayanan, R.,
Beuerman, R. W. et al. Rational
optimization of conformational effects
induced by hydrocarbon staples in
peptides and their binding interfaces.
Scientific Reports 3, 3451 (2013).
Computer simulations help design molecules with potential as a
cancer therapy

2
0
1
4

A
*
S
T
A
R

p
5
3

L
a
b
o
r
a
t
o
r
y
Our next step is to
further understand how
and why certain stapled
peptides are biologically
active whilst others
are not.
GENETICS &
DISEASE
Pathogenic fungi like Candida
albicans can cause oral, skin, nail
and genital infections. While expo-
sure to pathogenic fungi is gener-
ally not life-threatening, it can be
deadly to immunocompromised
patients with AIDS or cancer. A
variety of antifungal medications,
such as triazoles and polyenes, are
currently used for treating fungal
infections. The range of these
antifungal medications, however,
is extremely limited, with some
fungal species developing resist-
ance to these drugs.
Yi Yan Yang at the A*STAR
Nanotechnology in Singapore and
co-workers, in collaboration with
IBM Almaden Research Center in
the United States, have discovered
four cationic terephthalamide-
bisurea compounds with strong
antifungal activity, excellent
microbial selectivity and low host
toxicity
1
. These small molecules
can self-assemble into fibers, bind
fungal membranes and rupture the
cells of a variety of fungal species.
The results may expand the pos-
sibilities of medication for battling
drug-resistant fungal species.
Conformational analysis
revealed that the terephthalamide-
bisurea compounds have
a Z-shaped structure: the
terephthalamide sits in the
middle, urea groups on both
sides of the terephthalamide, and
cationic charges at both ends. The
researchers prepared compounds
with different spacers ethyl,
butyl, hexyl or benzyl amine
in-between the urea group and the
cationic charge.
When dissolved in water,
the terephthalamide-bisurea
compounds aggregate to form
fibers with lengths ranging from
a few hundred nanometers to
several micrometers. Some of
the compounds form fibers with
high f lexibility and others with
high rigidity.
The researchers evaluated
the antifungal activity of
their terephthalamide-bisurea
compounds against C. albicans.
They found that all of the cationic
compounds effectively inhibited
fungal growth, even when the
fungal concentration increased
from 10
2
to 10
5
colony-forming
units per milliliter.
The researchers believe that
the potent antifungal activity
is largely due to the forma-
tion of fibers with extremely
small diameters in the order
of 5 to 10 nanometers, which
facilitates the rupture of fungal
membranes. This is particularly
important because the fungal
membrane of C. albicans is
multilayered and has low negative
charges, explains Yang. It
also helps explain why cationic
terephthalamide-bisurea com-
pounds could easily penetrate the
fungal membrane.
The terephthalamide-bisurea
compounds also eradicated
clinically isolated drug-resistant
C. albicans. The compounds
prevent the development of drug
resistance by rupturing the fungal
membrane of C. albicans and
disrupting the biofilm (see image).
Additionally, cytotoxicity
tests showed that the cationic
terephthalamide-bisurea com-
pounds exhibit low toxicity
toward mammalian cells and in a
mouse model, revealing that the
compounds are relatively safe for
preventing and treating fungal
infections, says Yang.
1. Fukushima, K., Liu, S., Wu, H.,
Engler, A. C., Coady, D. J. et al.
Supramolecular high-aspect ratio
assemblies with strong antifungal
activity. Nature Communications 4,
2861 (2013).
Nanomaterials:
Antifungals go nano
Cationic small molecules hold great potential for preventing and treating
fungal infections
R
e
p
r
o
d
u
c
e
d
,
w
i
t
h

p
e
r
m
i
s
s
i
o
n
,
f
r
o
m

R
e
f

1
.

2
0
1
3

K
.
F
u
k
u
s
h
i
m
a

e
t

a
l
.
Scanning electron microscopy images of Candida albicans biofilm before (top) and after
(bottom) treatment with nanofibers produced by researchers at the A*STAR Institute of
Bioengineering and Nanotechnology. Animal studies demonstrated that the nanofibers
effectively eradicated C. albicans biofilm in a mouse fungal keratitis model.
TECHNOLOGIES
GENETICS &
DISEASE
Genomics:
The significance of silence
An enzyme that silences DNA activity may be crucially involved in health
and cancer
Understanding the action of Wip1, an
enzyme that controls DNA silencing, could
help explain both the development of
cancer and resistance to anticancer drugs.

S
v
i
s
i
o
/
i
S
t
o
c
k
/
T
h
i
n
k
s
t
o
c
k
Cancers can develop when the
complex molecular networks that
control the activity of DNA are
disrupted. Researchers from the
and Cell Biology in Singapore, led
by Dmitry Bulavin, have studied
Wip1, an enzyme central to these
molecular networks that may both
help to keep cells healthy but also
become part of the problem when
things go wrong
1
. The teams
findings suggest that this enzyme,
and its associated signaling
pathways, could be a target for
the development of new drugs to
combat some types of cancer. They
may also explain why some cancers
are resistant to drug therapy.
Bulavin and co-workers uncov-
ered the role of the Wip1 enzyme
when examining the deactivation
or silencing of regions of DNA by
DNA methylation. DNA methyla-
tion occurs when methyl (CH
3
)
groups are added to the cytosine
bases of DNA. This process is
known to be vital for the normal
control of gene activity as cells
and organisms develop. Changes
in DNA methylation patterns,
however, are also involved in the
development of many diseases,
including cancer.
Wip1 is known to influence
the activity of other molecules
by removing phosphate (PO
4
3
)
groups. However, enzymes
rarely act alone to regulate gene
activity they are components of
complex networks of interactions.
To clarify the Wip1 enzymes role
in health and disease, the team
looked at its direct and indirect
interactions with several other
molecules already implicated in
cancer, including DNA itself.
Working in mice and with
human cancer cells, the researchers
examined the effects of creating
Wip1 deficiency, as well as stimu-
lating the overexpression of the
gene that makes Wip1. They found
varied effects on more than a thou-
sand genes, including increases
and decreases in DNA silencing
when the normal activities of Wip1
were disrupted. Taken together,
the results suggest that the Wip1
enzyme plays an important role
in controlling DNA methylation
in tightly coiled and often
inactive regions of DNA
known as heterochromatin. The
enzymes DNA-silencing effects are
mediated through interaction with
two other proteins involved in the
onset of cancer, known as ATM
and BRCA1.
The teams results may be
particularly relevant to muta-
tions in primary breast cancers.
Ultimately, cancer develops and
evolves as a result of mutations
that contribute to these processes,
but the phenomenon is very
poorly understood, says Bulavin.
We show Wip1 is critical to
maintaining the integrity of the
genome, which provides new
avenues to understanding cancer
evolution and the mechanisms
responsible for developing anti-
cancer drug resistance.
1. Filipponi, F., Muller, J., Emelyanov, A.
& Bulavin, D. V. Wip1 controls global
heterochromatin silencing via
ATM/BRCA1-dependent DNA
methylation. Cancer Cell 24,
528541 (2013).
GENETICS &
DISEASE
Reprogramming a cell to express a
foreign gene is relatively straight-
forward. However, reprogramming
a cell to express multiple foreign
genes is far more complex, espe-
cially if trying to exert control over
the interplay between these genes.
Now, researchers led by Yuansheng
Yang of the A*STAR Bioprocessing
Technology Institute, Singapore,
have developed a strategy to finely
control the relative activity of
multiple genes in parallel
1
.
Some of the complexity of repro-
gramming lies in balancing propor-
tions: many proteins consist of
multiple component subunits, and
for the complex to assemble prop-
erly, each subunit must be manu-
factured in specific proportions.
Yangs strategy was to manipulate
an RNA sequence known as an
internal ribosome entry site (IRES).
Most messenger RNA (mRNA)
molecules represent the product of a
single gene that gets translated into
a single protein, but some viruses
produce mRNAs that contain
protein-coding regions from mul-
tiple genes separated by an IRES.
This makes the IRES a powerful
scientific tool for multigene expres-
sion. As genes linked by an IRES
are translated independently, we
can adjust their relative expression
by varying the strength of the
IRES, explains Yang. However,
the number of naturally available
IRESs is limited and can only
provide a narrow range of gene
expression. Yangs team decided to
expand their options by generating
a library of IRESs with a variety
of sequence mutations. Then, they
assessed how these variants affected
relative protein production when
inserted between two genes.
To test the different IRES
constructs, the researchers produced
antibodies immune proteins
formed of large heavy-chain
and smaller light-chain protein
components. The 24 variants they
generated varied considerably in
relative gene activity, ranging from
a 12 to 96 per cent reduction in the
production of light-chain proteins
relative to heavy-chain proteins.
These differences had a
profound effect on mature
antibody assembly. For instance,
producing the two components
in equal proportions resulted in
high levels of antibody production
but also yielded many undesirable
byproducts. However, an IRES
that increased relative light-chain
to heavy-chain production by
50 per cent virtually eliminated
those byproducts without a
meaningful impact on total
antibody production.
A United States biotech
company has already expressed
interest in licensing this strategy
for improved antibody production.
Meanwhile, Yangs group has
additional applications in mind. For
example, these sequences could help
manipulate glycosylation pathways,
a crucial mechanism for functional
modification of proteins that
depends on finely choreographed
activity from multiple enzymes.
This can critically impact the
biological activity, serum half-life
and immunogenicity of therapeutic
proteins, says Yang.
A balance of heavy-chain (red and blue) and light-chain (green and yellow) production is
essential for proper antibody assembly.
I
m
a
g
e

c
o
u
r
t
e
s
y

o
f

T
i
m
V
i
c
k
e
r
s
/
W
i
k
i
m
e
d
i
a

C
o
m
m
o
n
s
Genomics:
Putting production
in proportion
1. Koh, E. Y. C., Ho, S. C. L.,
Mariati, Song, Z., Bi, X., Bardor, M. &
Yang, Y. An internal ribosome entry
site (IRES) mutant library for tuning
expression level of multiple genes in
mammalian cells. PLoS ONE 8,
e82100 (2014).
A library of sequences that modify relative gene expression enables tighter
control over protein production
Our strategy can be
used to critically impact
the biological activity,
serum half-life and
immunogenicity of
therapeutic proteins.
TECHNOLOGIES
GENETICS &
DISEASE
Scientists have known for decades
that the protein p53 is critical
for safeguarding against genetic
damage and the onset of cancer.
Many tumors exhibit mutations
that undermine p53s ability to
regulate the genes responsible for
activating DNA repair and halting
cell division. While p53 also
appears to be an important con-
troller of other cellular processes,
identifying the additional genes that
p53 targets remains challenging.
Now, Ee Chee Ren and col-
leagues at the A*STAR Singapore
Immunology Network have
developed an assay that could
accelerate identification of the
genes that p53 regulates
1
. p53
binds target genes directly via a
family of closely related DNA
sequences known as p53 response
elements (p53REs). These are
short DNA sequences found in a
genes promoter region that bind
specific proteins that regulate
gene transcription (see image).
Prior searches for p53 target
genes were informed by only
limited knowledge of the
interaction between p53 and
DNA; these studies compared
changes in gene expression
in the presence or absence of
normal, or wild-type, p53, but
without validating the genes that
appeared to be affected or their
physical association with p53.
However, Ren and colleagues
were able to use p53RE-defining
sequence elements that they had
previously identified
2
to design
a bioluminescence-based assay
capable of accurately determining
whether genes are subject to
p53-mediated regulation.
Rens team assembled a
panel of 16 p53 proteins with
slight structural differences by
combing a database of DNA
sequences from cancer patients
and identifying sequence changes
that affected the amino-acid
makeup of p53. This allowed
them to distinguish variants
that exhibited similar p53RE
binding and gene activation
to wild-type p53 from those
that were nonfunctional.
Importantly, they found that
similarly classified p53 variants
exhibited the same DNA binding
and gene activation profiles as
each other, enabling the effects
of p53 to be determined from
multiple, parallel experiments.
We developed an efficient
screening approach that can
examine putative p53 target genes
based on the signature obtained
with these 16 variants, instead of
relying on only one test with just
the wild-type p53, explains Ren.
The teams assay proved far
more efficient at accurately
identifying known p53 target
genes than standard experimental
techniques, confirming its
effectiveness as a discovery tool.
The researchers demonstrated this
by using their panel to identify
nearly 600 new target genes of
p53 that affect diverse cellular
functions. Our results show that
the reach of p53 is far wider than
previously believed, and that
it participates in many normal
physiological processes of the
cell, says Ren.
Genetics:
Finding consensus on
target genes
A panel of p53 variants helps scientists condently identify genes controlled
by the master regulator protein
The interaction between p53 protein (top) and the p53 response element DNA sequence
of a target gene (DNA helix). Yellow arrows indicate the positions of several amino acid
sequence variations affecting the DNA-binding region of p53.
R
e
p
r
o
d
u
c
e
d
,
w
i
t
h

p
e
r
m
i
s
s
i
o
n
,
f
r
o
m

R
e
f
.
1

2
0
1
4

B
.
W
a
n
g

e
t

a
l
.
1. Wang, B., Niu, D., Lam, T. H.,
Xiao, Z. & Ren, E. C. Mapping the
p53 transcriptome universe
using p53 natural polymorphs.
Cell Death & Differentiation 21,
521532 (2014).
2. Wang, B., Xiao, Z. & Ren, E. C.
Redefining the p53 response
element. Proceedings of the
National Academy of Sciences USA
106, 1437314378 (2009).
GENETICS &
DISEASE
A computer-based model of cell
particle dynamics shows that
the lopsided torque produced by
forces within a cell may explain
the previously puzzling motion
and shape of rotating cell pairs
1
.
The model provides novel insights
that challenge current thinking
about the causes of developmental
abnormalities and of cancer and
other diseases linked to disrupted
cell rotation.
A crucial feature of the model
of key particle motions, developed
by Fong Yew Leong at the A*STAR
Computing in Singapore, is that
it spontaneously produces cells
with the same shape and rotational
characteristics as real cells.
During the development of
multicellular organisms, cells
rotate around each other in
a coordinated manner. This
rotation does not occur in some
cancer cells, which implies that
disrupting normal cell movement
may inf luence disease as well
as development.
Isolated cells growing on
f lat surfaces tend not to rotate.
However, two joined cells can
rotate spontaneously and continu-
ously, and will often develop a
sigmoidal or S-shaped interface
that resembles the yin and yang
symbol (see image). As the cells
rotate, they appear to moonwalk
around one another; each cell
moves in the opposite direction
to its protrusion into the other cell
and in the reverse manner to cells
moving on their own.
Leong modeled the cells in
two dimensions as an assembly of
cytoplasm particles surrounded
by a cell membrane. As cell
movement is driven by the
interaction of actin and myosin
protein filaments in the cells
cytoskeleton, Leong designed the
model to include the formation
and degradation of actin and
myosin chains attached to the
inner cell membrane.
The simulation showed the
interaction of actin and myosin
within the cell known
as undirected actomyosin
forcing which is powerful
enough to generate the shape and
movement of the cells. Crucially,
as Leong explains, forces that are
angled toward the cell membrane
lead to an unbalanced torque that
rotates the cell. He also considers
the spontaneous emergence of the
torque, due to the tilted forces,
to be the most significant insight
provided by his model.
The next step is to develop
a three-dimensional model that
explains cell cluster rotation
in vivo, rather than just on
a two-dimensional surface,
says Leong. He hopes future
iterations of the rather simple
current model will help explain
the real-life complexity of
the movements of multiple
cells and ultimately advance
approaches to addressing
developmental abnormalities and
diseases linked to the disruption
of cell movement.
Cell motion:
The yin and yang of
moonwalking cells
A particle dynamics model may provide insight into diseases and
deformities linked to disruptions in cell development
The cell cytoplasm (blue) interacts with the cell membrane (red) through actomyosin activity (silver rods), as simulated using the
particle dynamics model.

2
0
1
4

A
*
S
T
A
R

I
n
s
t
i
t
u
t
e

o
f

H
i
g
h

P
e
r
f
o
r
m
a
n
c
e

C
o
m
p
u
t
i
n
g
1. Leong, F. Y. Physical explanation
of coupled cell-cell rotational
behavior and interfacial
morphology: A particle dynamics
model. Biophysical Journal 105,
23012311 (2013).
TECHNOLOGIES
GENETICS &
DISEASE
The information contained within
a messenger RNA (mRNA)
transcript goes beyond the protein
recipe embedded in its sequence.
mRNAs consist of single strands
of nucleotides that can pair with
each other in the same way as
double-stranded DNA molecules.
The resulting secondary structures
help determine when and how the
encoded protein gets produced.
Researchers led by Yue Wan
of the A*STAR Genome Institute
of Singapore and Howard Chang
of Stanford University in the
United States have applied a
powerful experimental technique to
build a detailed map of secondary
structures for the entire human
mRNA transcriptome
1
.
Their Parallel Analysis of
RNA Structure (PARS) method
entails isolating the total mRNA
content of a biological sample,
then treating it with two different
enzymes that selectively cut single-
or double-stranded RNA segments.
By using sequencing technology to
map these cut sites, the researchers
could chart the secondary struc-
ture of each mRNA transcript.
The analysis uncovered some
interesting general features of
mRNA structure. For example,
protein-coding regions had
less structure than noncoding
regulatory sequences, particularly
in segments involved in splicing
an enzymatic process that
expands the number of proteins
encoded by a single transcript.
Additionally, nearly 10 per cent of
the mRNAs the team examined
assumed multiple secondary
structure arrangements, sug-
gesting that switching between
conformations plays an important
regulatory role.
The researchers also investigated
instances where mRNA structure
is affected by differences in the
genomic DNA sequence from
which it was transcribed. There is
extensive genetic variation between
individuals, explains Wan. To
understand the extent to which
this causes structural alterations in
humans, we performed a genome-
wide analysis in a family of three
individuals. This showed that
seemingly minor differences can
have a considerable impact: roughly
15 per cent of the single-nucleotide
sequence variations between
individuals caused structural
changes in an mRNA (see image).
Although other researchers
predicted that genome sequence dif-
ferences would have such an effect,
Wans findings represent the first
direct demonstration of the extent
of this phenomenon. We identified
over 1,900 nucleotide variants that
cause structural changes in the
human transcriptome far more
than anybody else has discovered
previously, he says.
In several instances, the
researchers found evidence that
these variant-associated changes may
impact gene regulation, including
protein production, and therefore
contribute to certain disease states.
This work was done on healthy
individuals, but our findings suggest
that some mutations may cause
disease by altering gene regulation,
says Wan. Future work could
compare diseased with normal
tissues to identify and characterize
structure-changing mutations.
Genomics:
Structure informs function
Genomic diferences between individuals can change the physical
organization of RNA transcripts, afecting their expression
G
G
A
A
G
A
A
C
U
U
U
C
U
G
C
C
C
C
G
G
C
U
U
U
G
C
A
G
G
A
U
G
A
A
A A
G
C
U
U
U
C
C
U
G
C
U
U
G
G
C
A
G
U
U
A
U
U
C
U
U
C
C G
G
A
A
G
A
A
C
U
U
U
C
U
G
C
C
C
C
G
G
C
U
U
U
G
C
A
G
A
A
U
G A
A
A
A
G
C
U
U
U
C
C
U
G
C
U
U
G
G
C
A
G
U
U
A
U
U
C
U
U
C
C
Father allele- G
Mother allele- A
Single-nucleotide differences in genomic sequence between two individuals, such as the
substitution of guanine (G) with adenine (A), can considerably alter the structure of the
resulting mRNA transcript.

2
0
1
4

A
*
S
T
A
R

G
e
n
o
m
e

I
n
s
t
i
t
u
t
e

o
f

S
i
n
g
a
p
o
r
e
1. Wan, Y., Qu, K., Zhang, Q. C.,
Flynn, R. A., Manor, O. et al.
Landscape and variation of RNA
secondary structure across the
human transcriptome. Nature 505,
706709 (2014).
GENETICS &
DISEASE
Being able to view tumor blood ves-
sels without surgery or potent dyes
can improve our understanding of
the environment in which a tumor
grows. Now, a team of researchers,
including Chang-Tong Yang from
the A*STAR Singapore Bioimaging
Consortium, have developed a
contrast agent that selectively labels
the tumor blood vessel network
for use in noninvasive tumor
imaging studies
1
.
Assessing the extent of the
blood vessel network within a
tumor enables doctors to deter-
mine how far along the cancer has
progressed and to predict whether
a tumor will respond to intrave-
nous chemotherapy drugs through
modeling the drugs potential
path and absorbance. While
noninvasive imaging techniques
can help visualize all blood vessels
in patients, visualizing the blood
vessels of a tumor requires a
contrast agent that can selectively
label the tumor vessels. Such
substances concentrate within
tumors by binding to molecules
that are more prevalent in tumors
than in normal tissues.
The researchers developed a
contrast agent called Gd(DO3A-
Lys) that has an optimal half-life,
meaning it persists in the blood
long enough to be useful as an
imaging aid. Contrast agents
that strongly bind blood albumin
increase the visibility of large blood
vessels and are already in clinical
use, explains research co-leader
Edward George Robins. However,
Gd(DO3A-Lys) only weakly binds
albumin, enabling the contrast
agent to specifically target and
visualize areas that contain tumors
and their microvasculature.
Contrast agents eventually
leave the body when filtered
through the kidneys. Promisingly,
the researchers were able to show
that Gd(DO3A-Lys) has low
toxicity to kidney cells growing in
culture. In addition, mice given
a single dose of Gd(DO3A-Lys)
showed no kidney toxicity at both
low and high doses, although the
high-dose treatment did result in
some damage to the liver the
organ responsible for breaking
down drugs.
After administering Gd(DO3A-
Lys) to tumor-bearing mice, Yang
and colleagues used contrast-
enhanced magnetic resonance
imaging (CE-MRI) to demonstrate
that the contrast agent selectively
localized to the site of the tumor.
Importantly, as only a minimal
amount leaked from the blood
vessels into the tumor itself,
Gd(DO3A-Lys) appeared to be
retained within the tumors blood
vessel network, allowing the fine
structures of the tumor vessels to
be clearly imaged.
While the teams results are
encouraging, clinical use of
Gd(DO3A-Lys) is still a long way
off. Further study is needed to
assess the sensitivity of Gd(DO3A-
Lys) to detect changes in the tumor
vasculature and potential interac-
tions with tumor treatments such as
anti-angiogenesis drugs.
Cancer:
A view of tumor vessels
An agent to selectively label tumor blood vessels during noninvasive tumor
imaging could help determine a likely response to chemotherapy
Improved imaging techniques for tumor blood vessel networks will help improve the targeting of intravenous chemotherapy.

B
l
u
e
P
l
a
n
e
t
E
a
r
t
h
/
i
S
t
o
c
k
/
T
h
i
n
k
s
t
o
c
k
1. Yang, C.-T., Chandrasekharan, P.,
He, T., Poh, Z., Raju, A., Chuang, K.-H.,
& Robins, E. G. An intravascular MRI
contrast agent based on Gd(DO3A-
Lys) for tumor angiography.
Biomaterials 35, 327336 (2014).
TECHNOLOGIES
GENETICS &
DISEASE
Dysfunctions in cilia, tiny
hair-like structures that protrude
from the surface of cells, are
responsible for a number of
human diseases. However the
genes involved in making cilia
have remained largely elusive. In
the first comprehensive analysis
of its kind, researchers from the
and Cell Biology in Singapore
have identified hundreds of genes
involved in the proper formation
of a particular type of cilia that
help to remove mucus and dirt
from the lungs
1
.
This collection of genes will
be invaluable for our under-
standing of how cilia are made,
says Sudipto Roy, a developmental
biologist and senior author of the
study. More importantly, it will
greatly facilitate the diagnosis of
cilia-based diseases.
Cilia come in two forms:
primary, or non-motile, cilia that
serve as environmental sensors for
cells throughout the body, and
motile cilia, which constantly
beat to sweep debris from the
middle ear and respiratory tract
(see image). Mutations that affect
the function of motile cilia play
a role in genetic disorders like
primary ciliary dyskinesia, which
affects the lungs and other sites
where such cilia are present.
Previously, Roy and colleagues
found that a transcription factor
called forkhead box protein J1
(Foxj1) is the master regulator of
motile cilia development
2
. The
researchers demonstrated that
Foxj1 controls the expression
of numerous genes involved in
the production of these moving,
microscopic protrusions.
However, they were unsure as to
exactly which genes were impli-
cated in the process of making
functional motile cilia.
Roys team, guided by senior
research fellow Semil Choksi,
therefore decided to perform
a systematic analysis of all the
genes in the zebrafish, a common
model of cilia development. They
discovered more than 500 genes
with mammalian counterparts
that are activated by Foxj1,
the majority of which had not
previously been associated with
cilia production.
The researchers randomly
selected 50 of these genes
for functional studies. They
artificially boosted expression
levels of each gene by injecting
lab-made RNA into zebrafish
embryos, and found that around
30% of the proteins encoded
by the genes localized to the
motile cilia. The researchers also
inactivated each of these 50 genes
in turn using morpholinos, a
standard knockdown tool. In this
way, they showed that more than
60% of the genes were needed
for the proper differentiation or
functioning of motile cilia.
According to Choksi, the
teams collection of cilia-
associated genes is set to help
researchers identify previously
unknown mutations behind
cilia disorders in patients and
ultimately, perhaps, new therapies.
Cell biology:
The genes behind the guardians
of the airways
A genetic screen in zebrash identies genes implicated in the development
of cell structures linked to respiratory disease in humans
Cells from the human airway covered with hundreds of motile cilia (green).

2
0
1
4

A
*
S
T
A
R

I
n
s
t
i
t
u
t
e

o
f

M
o
l
e
c
u
l
a
r

a
n
d

C
e
l
l

B
i
o
l
o
g
y
1. Choksi, S. P., Babu, D., Lau, D.,
Yu, X. & Roy, S. Systematic
discovery of novel ciliary genes
through functional genomics in
the zebrafish. Development 141,
34103419 (2014).
2. Yu, X., Ng, C. P., Habacher, H. &
Roy, S. Foxj1 transcription factors
are master regulators of the motile
ciliogenic program. Nature Genetics
40, 14451453 (2008).
Research Highlights
PHYSICAL & LIFE
SCIENCE TECHNOLOGIES
TECHNOLOGIES
A new metal carbonyl-based assay has
potential as a clinical diagnostic tool and
offers several advantages over current
monitoring of sugars in urine samples.

m
o
o
d
b
o
a
r
d
/
T
h
i
n
k
s
t
o
c
k
Medical diagnostics:
Sweet sensing
Diabetes is a major health problem
that affects 371 million people
worldwide, and a high blood glu-
cose level is one of many complica-
tions associated with diabetes and
pre-diabetes. Malini Olivo at the
A*STAR Singapore Bioimaging
Consortium and co-workers have
now developed a highly sensitive
and specific surface-enhanced
Raman scattering (SERS)-based
assay to detect glucose in urine
1
.
SERS-based assays represent
a simple and convenient method
to monitor glucose levels in urine.
The technology uses organo-
metallic compounds as sugar
receptors for sensing glucose and a
goldsilver substrate for enhancing
signals from these receptors. Most
such sugar receptors, however, have
a low detection limit and speci-
ficity. They often require samples
to be purified prior to testing and
may not be able to distinguish
glucose from similar sugars such as
fructose and galactose. As a result,
SERS-based assays are not yet an
accessible and dependable solution
for monitoring glucose levels in
patients with diabetes.
Searching for a more effective
test, Olivo and team developed
an assay which uses triosmium
carbonyl conjugates as sugar
receptors and an in-house fabri-
cated material bimetallic film
over nanospheres (BMFON) as
the signal-enhancing substrate.
They found that the metal
carbonyl conjugates selectively
bind simple sugars and produce a
unique absorption peak for glucose
at 2111 centimeters
1
, which lies in
the so-called silent mid-infrared
region of the SERS spectrum
between 18002200 centimeters
1
.
The intensity of the peak varies
with glucose concentration and
can be used to measure glucose
quantities. In addition, BMFON
enhances the signal intensity
to extend the limit of detection
to 0.1 millimoles of glucose
per liter to cover a range that
includes the physiological glucose
concentration of 5 millimoles
per liter.
Using their newly developed
assay, the researchers could
determine the glucose level in
urine samples spiked with a
standard glucose solution. The
new assay has three key advan-
tages: only a low sample volume
is required; there is no need for
the sample to be purified; and
the assays sugar receptors do not
need to be conjugated to SERS-
active nanoparticles.
Our work is unique in that
we used a metal carbonyl probe to
access an uncluttered region in the
SERS spectrum, which most other
organic probes cannot provide,
says Olivo. By coupling this probe
to a SERS-based assay, we achieved
a high selectivity and sensitivity
in the detection of glucose over
other sugars.
These advantages should mean
that this concept for a glucose assay
can be developed into a clinical
diagnostic tool, says Olivo.
A highly sensitive and specic biological assay requires only a tiny sample
volume to monitor glucose levels in patients with diabetes
1. Kong, K. V., Lam, Z., Lau, W. K. O.,
Leong, W. K. & Olivo, M. A transition
metal carbonyl probe for use in a
highly specific and sensitive SERS-
based assay for glucose. Journal
of the American Chemical Society
135, 1802818031 (2013).
TECHNOLOGIES
1. Cheng, M.-Y., Je, M., Tan, K. L.,
Tan, E. L., Lim, R. et al. A low-
profile three-dimensional
neural probe array using a
silicon lead transfer structure.
Journal of Micromechanics and
Microengineering 23,
095013 (2013).
Neural probe arrays are expected
to significantly benefit the lives
of amputees and people affected
by spinal cord injuries or severe
neuromotor diseases. By providing
a direct route of communication
between the brain and artificial
limbs, these arrays record and
stimulate neurons in the cer-
ebral cortex.
The need for neural probe
arrays that are compact, reliable
and deliver high performance
has prompted researchers to use
microfabrication techniques to
manufacture probe arrays. Now,
a team led by Ming-Yuan Cheng
Microelectronics, Singapore, has
developed a three-dimensional
probe array for chronic and long-
term implantation in the brain
1
.
This array is compact enough to
freely float along with the brain
when implanted on the cortex.
The neural probe array needs to
be implanted in the subarachnoid
space of the brain, a narrow region
of 12.5 millimeters in depth that
lies between the pia mater and dura
mater brain meninges. A high-
profile array may touch the skull
and damage the tissue when relative
micromotions occur between the
brain and the probes, explains
Cheng. To avoid this problem, the
array should be as thin as possible.
Existing approaches produce
low-profile arrays using microscopic
electrodes machined from a silicon
substrate. These approaches, how-
ever, restrict the maximum probe
length to the thickness of the sub-
strate and the number of recording
electrodes. Other methods generate
three-dimensional arrays from
silicon-supported two-dimensional
probes. Complex and expensive to
fabricate, these arrays are too bulky
because the silicon support also
incorporates the application-specific
integrated circuit (ASIC) chip for
neural recording.
Cheng and colleagues fabricated
two-dimensional probes and
inserted them into a thin slotted
silicon platform for assembly (see
image). To produce a three-dimen-
sional probe array, they joined this
assembly to the recording chip.
Instead of being aligned, however,
the team found that the contacts of
the probe electrodes and recording
chip were orthogonally arranged
with respect to each other, resulting
in mismatched planes.
To solve this issue, the team
manufactured a silicon-based con-
nector, or interposer, that electri-
cally linked these components,
says Cheng. This innovative
microassembly effectively controls
the final height of the array to
within 750 micrometers.
Compared with commercial
neural probes, the new array
exhibited competitive electrical
properties, including electrode
impedance. Moreover, biocompat-
ibility tests showed that the pres-
ence of array components did not
rupture cell membranes or suppress
cell growth. The team is currently
refining their approach to integrate
the array with a wireless recording
chip and make the assembly
fully implantable.
Neural probes:
Slimmed down for a
better fit
A thinner probe array that uses a silicon-based microstructure could
underpin safer neural implants
The compact neural probe array consists of a three-dimensional probe array, a custom 100-channel neural recording chip and a flexible
polyimide polymer cable.

2
0
1
4

A
*
S
T
A
R

I
n
s
t
i
t
u
t
e

o
f

M
i
c
r
o
e
l
e
c
t
r
o
n
i
c
s
TECHNOLOGIES
Cataracts are the main cause of
visual impairment worldwide and
account for more than 50 per cent
of blindness in developing countries.
Diagnosis of the condition generally
involves a labor-intensive method
of ophthalmic examination and
grading by comparing the indi-
viduals lens against a set of standard
reference photographs (see image).
Now, Yanwu Xu and co-workers
from the A*STAR Institute for
Infocomm Research, together with
researchers in Singapore and China,
have developed a system to grade
the severity of cataracts automati-
cally using slit-lamp photographs
1
.
The technique offers potential for
improved clinical management of
cataracts and may be extended to
other eye conditions in the future.
Cataracts are categorized as
nuclear, cortical or posterior,
according to their location and
appearance. The researchers
studied nuclear cataracts, the most
common type of cataract, which
involve clouding, or opacification,
of the central part of the lens. They
analyzed a database of 5378 images
taken using slit-lamp illumination,
the standard method for examining
and photographing cataracts.
From this analysis, the research
team developed a computer-based
system in which each image of a
lens is divided into three sections
nucleus, anterior cortex and
posterior cortex and visual
features are extracted from a grid
of overlapping patches for analysis.
The software uses a statistical
process, known as group sparsity
regression, to determine each
cataracts grade on a standard
scoring system. The distinguishing
aspect of the group sparsity regres-
sion step is that it selects just a
few groups of features of intensity
and texture in each image, which
reduces the influence of random
noise and increases the speed
of computation.
While not the first attempt
to grade cataracts automatically,
the researchers found that their
system outperforms existing
methods by more closely matching
the results of manual grading by
experienced clinicians.
Manual assessments can be
subjective, time-consuming and
costly, says Xu. Our automated
system offers the prospect of
greater objectivity and efficiency
in diagnosing and monitoring this
major condition in huge numbers
of people. In addition to the initial
diagnosis it could also help doctors
to monitor the progression and
treatment of cataracts.
The researchers now hope
to refine their software and to
explore the possibility of applying
it to other eye conditions such as
pathological myopia, glaucoma and
age-related macular degeneration.
Their ultimate aim is to develop
a more comprehensive com-
mercialized system.
The computation is not com-
plex, explains Xu. It could form
an ocular screening platform based
on cloud computing technology that
could also provide a remote service.
Standard slit-lamp photographs of a normal eye (left) and a cataract (right) used in
manual assessments of cataracts.
Ophthalmology:
Grading cataracts automatically
1. Xu, Y., Gao, X., Lin, S., Wong, D. W. K.,
Liu, J. et al. Automatic grading
of nuclear cataracts from slit-
lamp lens images using group
sparsity regression. Medical Image
Computing and Computer-Assisted
Intervention MICCAI 2013, Lecture
Notes in Computer Science 8150,
468475 (2013).
Software to grade cataracts automatically could lead to improved
management and allow for remote assessment of the condition
R
e
p
r
o
d
u
c
e
d
,
w
i
t
h

p
e
r
m
i
s
s
i
o
n
,
f
r
o
m

R
e
f

1
.

2
0
1
3

S
p
r
i
n
g
e
r
-
V
e
r
l
a
g

B
e
r
l
i
n

H
e
i
d
e
l
b
e
r
g
Our automated system
ofers the prospect of
greater objectivity and
efciency in diagnosing
and monitoring this
major condition in huge
numbers of people. In
addition to the initial
diagnosis it could also
help doctors to monitor
the progression and
treatment of cataracts.
TECHNOLOGIES
Dwindling crude oil reserves,
accompanied by rising prices and
environmental concerns, have led
to increased interest in the use of
renewable fuels. Biofuels produced
from waste agricultural or forestry
material are particularly desirable
because they avoid diverting
resources from the production of
food crops.
Oils produced by high-
temperature treatment of these
waste materials, however, contain
a large amount of oxygenated
compounds that result in undesir-
able properties such as high
viscosity and corrosiveness. Now,
Jie Chang, Armando Borgna and
co-workers from the A*STAR
Institute of Chemical and
Engineering Sciences in Singapore
describe a series of catalysts
that might be used to upgrade
these oils by removing the
undesirable oxygen-containing
functional groups
1
. Using the
compound guaiacol as a model for
oxygenated bio-derived oils, the
researchers found that the most
promising catalysts for guaiacol
deoxygenation are comprised
of molybdenum metal on a
carbon support.
The diversity of sources of
waste biomass means that there
is great variability in the content
of the bio-oils produced by the
initial heat treatment, which is
itself the subject of much research.
Guaiacol provides, in a single
and easily available compound,
the types of oxygen-containing
functional groups that typically
need to be removed.
Catalysts for the related
process of desulfurization are
widely used in petroleum refin-
eries to produce cleaner fuels, but
they are not optimized for deoxy-
genation. The desulfurization
catalysts are well developed and
understood because of extensive
research into the mechanisms
by which they work, explains
Chang. We are using guaiacol
as a model compound to develop
a similar level of understanding
for deoxygenation.
The best catalysts identified
by the researchers show complete
conversion of guaiacol and over
80 per cent selectivity to the
desired hydrocarbon products
within minutes.
Chang and co-workers
undertook a detailed study of the
structure of the catalysts before
and during the reaction, as well as
of the catalysts that were deac-
tivated. They also attempted to
identify the reaction process in
particular, the types of oxygen-
containing functional groups that
react first and whether this affects
the performance of the catalyst.
While catalyst selectivity is
critical, other factors such as the
activity and stability of the cata-
lyst will prove equally important
because of their impact on the
economics of the overall process.
There is a long way to go before
this complete biomass-to-fuel
process can become commercial,
says Chang. Also, we hope
to develop the selectivity even
further so that it becomes useful
for developing fine chemicals as
well as fuels.
1. Chang, J., Danuthai, T.,
Dewiyanti, S., Wang, C. & Borgna,
A. Hydrodeoxygenation of guaiacol
over carbon-supported metal
catalysts. ChemCatChem 5,
30413049 (2013).
Biofuels:
Catalytic upgrade
New catalysts to remove oxygenated compounds from bio-derived oils may
lead to better and cheaper renewable biofuels
M
a
t
a
u
w
/
i
S
t
o
c
k
/
T
h
i
n
k
s
t
o
c
k
The deoxygenation of biomass such as corn stover or forestry waste could be the key to
development of viable biofuels.
We hope to develop the
selectivity even further so
that the catalyst becomes
useful for developing ne
chemicals as well as fuels.
TECHNOLOGIES
Biomimetic materials:
Keeping in contact
Transparent polymeric lms with near-uniform, continuous nanoprotrusions
show high water pinning abilities
A water droplet adheres to a pat-
terned polycarbonate film even when
held vertically.

2
0
1
4

A
*
S
T
A
R

I
n
s
t
i
t
u
t
e

o
f

M
a
t
e
r
i
a
l
s

R
e
s
e
a
r
c
h

a
n
d

E
n
g
i
n
e
e
r
i
n
g
A*STAR researchers have used
nanoimprinting methods to make
patterned polymeric films with
surface topography inspired by that
of a rose petal, producing a range of
transparent films with high water
pinning forces
1
.
A surface to which a water
droplet adheres, even when it is
turned upside down, is described
as having strong water pinning
characteristics. A rose petal and
a lotus leaf are both superhy-
drophobic, yet dissimilarities in
their water pinning properties
cause a water droplet to stick to
a rose petal but roll off a lotus
leaf. The two leaf types differ
in their micro- and nanoscale
surface topography and it is these
topographical details that alter the
water pinning force. The rose petal
has almost uniformly distributed,
conical-shaped microscale
protrusions with nanoscale folds
on these protrusions, while the
lotus leaf has randomly distributed
microscale protrusions.
The imprinted surfaces devel-
oped by Jaslyn Law and colleagues
Materials Research and Engineering
and the Singapore University
of Technology and Design have
uniformly distributed patterns of
nanoscale protrusions that are either
conical or parabolic in shape. The
researchers found that the water
pinning forces on these continu-
ously patterned surfaces were much
greater than on non-patterned
surfaces and surfaces composed of
isolated nanopillared structures or
nanoscale gratings. They could then
achieve high water pinning forces
by patterning the nanoprotrusions
onto polymeric films with a range of
different non-patterned hydropho-
bicities, including polycarbonate,
poly(methyl methacrylate) and
polydimethylsiloxane (see image).
Other methods that recreate
the water pinning effect have used
actual rose petals as the mold,
but unless special care is taken,
there are likely to be defects and
inconsistencies in the recreated pat-
tern, says co-author Andrew Ng.
While bottom-up approaches for
making patterns for example,
laser ablation, liquid flame spray or
chemical vapor deposition are
more consistent, these methods are
limited in the types of patterns that
can be used and the scale at which a
substrate can be patterned.
In contrast, nanoimprinting
methods are capable of fabricating
versatile and large-scale surfaces, and
can be combined with roll-to-roll
techniques, hence potentially ena-
bling more commercial applications.
The patterned polycarbonate
surfaces were also shown to reduce
the coffee-ring effect: the unevenly
deposited film left behind upon
the evaporation of a solute-laden
droplet. This mitigation of the
coffee-ring effect may assist
microfluidic technologies and, more
generally, the patterned surfaces
could be used in arid regions for
dew collection or in anti-drip
applications such as in greenhouses.
1. Law, J. B. K., Ng, A. M. H., He, A. Y. &
Low, H. Y. Bioinspired ultrahigh
water pinning nanostructures.
Langmuir 30, 325331 (2014).
TECHNOLOGIES
Cancer drugs can be modified to
specifically target tumor sites to
help personalize cancer treatment.
And while it is relatively easy to
determine if the drugs have been
delivered to the correct location, it
is more difficult to monitor their
therapeutic success. Now, a team
led by Bin Liu from the A*STAR
and Engineering in Singapore, in
collaboration with Ben Zhong Tang
at Hong Kong University of Science
and Technology, has developed
an anticancer drug with an
inbuilt mechanism that shows if it
is working
1
.
Platinum-based drugs are effec-
tive against many cancers, killing
cells by triggering cellular suicide,
or apoptosis. These drugs can,
however, have severe side effects.
Nontoxic forms can be modified,
as a type of prodrug, to convert to
their toxic form only after entering
the targeted tumor cells, so as not to
harm noncancerous cells.
Liu and colleagues went one step
further by modifying a platinum-
based prodrug to not only target
tumor cells effectively, but also
show whether or not it was having
the desired effect. According to Liu,
this added feature could be crucial
for improving cancer treatment.
Early evaluation of a patients
response to a specific cancer
therapy is important in clinical
applications because it can mini-
mize the duration of ineffective
courses, explains Liu. The
effectiveness of cancer treatment
is commonly evaluated using
magnetic resonance imaging to
measure the tumor size, but this is
unsatisfactory as the change in size
is not obvious at the early stages
of therapy.
In their new system, Liu and her
team included an apoptosis sensor
that is released when the prodrug
converts to its toxic form inside the
targeted tumor cells. The toxic form
triggers cell apoptosis and activates
an enzyme called caspase 3, which
then cleaves the apoptosis sensor
and causes it to fluoresce green.
This provides a visual signal that
the drug is killing the cells.
Liu and her colleagues tested
the mechanism by treating
cultured cancer cells with the
modified platinum prodrug. They
observed a gradual increase in
fluorescence in the cancer cells,
reaching a maximum level of
fluorescence six hours after treat-
ment. Noncancerous cells were not
affected in the same way, further
proving the effectiveness of their
targeting mechanism.
Such noninvasive and real-time
imaging of drug-induced apoptosis
could be used to evaluate the
therapeutic response to a specific
anticancer drug at an early stage,
explains Liu. Our system
can simultaneously deliver the
therapeutic drugs and noninvasively
evaluate the therapeutic responses
in situ.
Cancer:
Modified drug gives a
green light for its own success
A modied anticancer drug can simultaneously target tumor sites and show
whether or not it is working
A modified platinum-based prodrug targets cancer cells and triggers cell death, or apoptosis, causing apoptosis sensors to
fluoresce green.

U
g
r
e
e
n
/
i
S
t
o
c
k
/
T
h
i
n
k
s
t
o
c
k
1. Yuan, Y., Kwok, R. T. K., Tang, B. Z.,
Liu, B. Targeted theranostic
platinum(IV) prodrug with a built-in
aggregation-induced emission light-
up apoptosis sensor for noninvasive
early evaluation of its therapeutic
responses in situ. Journal of the
American Chemical Society 136,
25462554 (2014).
TECHNOLOGIES
Platinum-based chemotherapy
drugs are commonly used to treat
a wide variety of solid tumors,
including cancers affecting the
breast, colon and lung. However,
only a small amount of these
anticancer drugs typically reaches
the target organ system. This
inefficiency not only reduces the
efficacy of the drug; it also leads
to severe side effects, ranging
from nausea to kidney toxicity
or deafness.
Charlotte Hauser and her
co-workers at the A*STAR
have now developed a platform for
the localized and sustained release
of platinum-based anticancer
therapeutics that overcomes
many of these limitations.
The research teams novel gel
formulation a combination of
specialized peptides and the drug
oxaliplatin led to dramatic
growth inhibition when injected
directly into the breast tumors of
mice. In addition, the gel treat-
ment produced a better tolerance
profile than standard oxaliplatin
drug therapy
1
.
Compared to the free drug,
our injectable drug-loaded
conjugate is just as effective in
inhibiting tumor growth, but with
lower systemic toxicity and higher
localization in the target tissue,
says Hauser.
Hausers team started with
a unique class of ultrashort
peptides that have the ability
to spontaneously self-assemble
and form hydrogels. The
researchers attached oxaliplatin
to the peptides using a technique
known as click chemistry, which
enables the synthesis of complex
molecules through the joining
of multiple attached parts (see
image). The resulting oxaliplatin-
peptide hydrogels proved highly
lethal against two human cell
lines derived from cervical
cancer and colon cancer tissues,
respectively. Laboratory analyses
showed that the hydrogels bound
to the DNA of the cancer cells,
arresting their replication cycle
just as free oxaliplatin does.
Furthermore, the whole construct
was biocompatible and generally
non-immunogenic.
The researchers compared
their hydrogel head-to-head with
unmodified oxaliplatin in a breast
cancer mouse model by injecting
the drugs locally into the mouse
tumors. Where the drug was
delivered as a hydrogel, they
documented higher rates of drug
accumulation in the tumor but
lower levels of drug toxicity in the
kidneys and livers compared
to the free oxaliplatin control.
As a result, the mice given the
hydrogel maintained more body
weight, which can be used as
a surrogate measure of their
overall health.
This combination product
could serve as a tissue replace-
ment device for the controlled
release of important drugs that
require localized and injectable
treatments, says Hauser. We are
exploring if this general approach
can be utilized to attach a variety
of other bioactive molecules.
Drug delivery:
A click toward localized
chemotherapy
Click chemistry produces a hydrogel with less toxicity and greater tissue
localization in a mouse cancer model
N
3
N
3
N
3
Click
Self-assembly
The oxaliplatin-peptide conjugate, formed using click chemistry, self-assembles into a
hydrogel (bottom left) that can be injected for localized drug delivery.
R
e
p
r
o
d
u
c
e
d
,
w
i
t
h

p
e
r
m
i
s
s
i
o
n
,
f
r
o
m

R
e
f
.
1

2
0
1
3

R
o
y
a
l

S
o
c
i
e
t
y

o
f

C
h
e
m
i
s
t
r
y
1. Reithofer, M. R., Chan, K.-H.,
Lakshmanan, A., Lam, D. H.,
Mishra, A. et al. Ligation of
anti-cancer drugs to self-
assembling ultrashort peptides
by click chemistry for localized
therapy. Chemical Science 5,
625630 (2014).
TECHNOLOGIES
In order to determine the best
treatment for patients affected by
cancer, it is crucial for physicians
to identify how the disease is
spreading via lymph nodes in the
body a process known as metas-
tasis. Progression of the disease is
currently monitored by dissecting
lymph nodes during surgery and
subsequently performing biopsies.
However, using a more sensitive and
accurate method that is less invasive
based on optical imaging technolo-
gies to visualize disease progression
in situ could further improve cancer
patient diagnoses and limit the
time that they are required to spend
in surgery.
Now, Young-Tae Chang and co-
workers at the A*STAR Singapore
Bioimaging Consortium, together
with Jung Sun Yoo and researchers
at the National University of
Singapore, have developed a novel
fluorescent probe that is capable
of infiltrating lymph nodes and
highlighting cancer progression
1
.
The probe could potentially be used
to provide information for making
rapid diagnoses during surgery.
Immune cells in lymph nodes,
such as macrophages, have a
novel role in disease progression,
explains Yoo. We wanted to find a
macrophage-targeting fluorescent
probe that could distinguish
metastasized lymph nodes from
inflamed or normal lymph nodes.
The team sifted through many
libraries of fluorochromes small
nontoxic organic molecules suitable
for intraoperative imaging while
searching for those that could
specifically stain macrophages.
Flow cytometry and testing with
human blood samples allowed the
researchers to select the best fit for
their fluorescent probe.
We then used the probe for
in vivo mouse imaging, explains
Yoo. This tested whether the probe
could pinpoint the sentinel lymph
node the first node to drain a
tumor and potentially initiate cancer
spreading through the body. Fol-
lowing injection, the teams probe
immediately accumulated in the
sentinel lymph node of the mouse,
with bright signals in inflamed
nodes and less bright signals in
metastasized nodes (see image).
The new probe has several
potential biomedical applications.
Researchers could use it to investi-
gate the behavior of macrophages in
lymph nodes in vivo. Surgeons could
obtain real-time information on
the infiltration of tumors and how
far the disease has spread while a
patient is still in surgery, eliminating
the need for pre-operative biopsies
and minimizing the potential for
further surgical procedures.
We also hope to develop
macrophage probes with near-
infrared for better depth penetra-
tion, as well as subset-specific agents
capable of differentiating between
good and bad macrophages in
many different diseases, states Yoo.
Subtype-specific fluorochromes
will have a high impact on the
future of clinical imaging.
Cancer surgery:
Lighting up lymph nodes
Advances in uorescent probe technology could allow real-time visualization
of cancer progression during surgery
The metastasized lymph node (top) of a mouse injected with the probe shows lower
fluorescence signal intensity than a normal lymph node (bottom).

2
0
1
4

J
u
n
g

S
u
n

Y
o
o
,
N
a
t
i
o
n
a
l

U
n
i
v
e
r
s
i
t
y

o
f

S
i
n
g
a
p
o
r
e
1. Yoo, J. S., Lee, S.-C., Jow, Z. Y.,
Koh, P. Y. X. & Chang, Y.-T.
A macrophage-specific fluorescent
probe for intraoperative lymph
node staging. Cancer Research 74,
4455 (2014).
We also hope to
develop macrophage
probes with near-
infrared for better depth
penetration, as well as
subset-specic agents
capable of diferentiating
between good and bad
macrophages in many
diferent diseases.
TECHNOLOGIES
Joint injuries often fail to
mend properly when not given
assistance. In particular, cartilage
exhibits a poor capacity for self-
repair. It is possible to stimulate
regeneration by implanting
synthetic scaffolds loaded with
cartilage-forming cells (known
as chondrocytes) at the site of an
injury. However, the quality of the
repair depends on the extent to
which healthy cartilage can form
on a given scaffold.
Now, researchers led by Moto-
ichi Kurisawa of the A*STAR
have devised an optimized
formula to encourage proper
regrowth of cartilage
1
. Kurisawas
group has worked extensively with
polymers known as hydrogels (see
image), whose stiffness strongly
depends on the conditions under
which they are synthesized. This
property enabled the researchers
to explore the extent to which the
hydrogel stiffness affects cartilage
formation. Research indicates
that cells that have adhered to a
solid substrate are able to sense
mechanical stimuli, and that
this can affect many important
physiological processes in a way
similar to biochemical signals,
says Kurisawa.
Ideally, joints should be
enriched in smooth hyaline
cartilage rather than tough
and dense fibrocartilage, and
the researchers suspected that
stiffer hydrogel scaffolds might
promote increased fibrocartilage
formation. To test this idea, they
synthesized three different kinds
of gelatinhydroxyphenylpropi-
onic acid (GtnHPA) hydrogels,
which they classified as having
low, medium or high stiffness.
Chondrocytes generally survived
well and proliferated when seeded
onto all three scaffolds, but they
produced cartilages of consider-
ably different qualities.
Chondrocytes cultivated on
GtnHPA-high produced the
highest density of cartilage, but
this tended to be primarily fibrous
in nature, whereas cells grown on
GtnHPA-low produced only low
levels of cartilage. However, the
medium-stiffness scaffold proved
well suited to the production
of hyaline cartilage, facilitating
efficient tissue repair after implan-
tation in a rabbit model of joint
injury. After four weeks, these
animals showed robust hyaline
cartilage growth; in contrast,
recipients of low- and high-
stiffness scaffolds exhibited only
limited repair or dense fibrous
tissue formation, respectively.
Importantly, the GtnHPA-
medium scaffold also readily
dissolved after implantation.
Our findings highlight the
importance of incorporating
considerations of hydrogel stiff-
ness into the design of scaffolds
intended for cartilage tissue
engineering, says Kurisawa. Based
on the optimal middle ground
identified in this study, he and his
colleagues are now hopeful that
they can demonstrate the same
level of safety and efficacy in large
animal models as the next step
toward moving to the clinic.
Regenerative medicine:
Finding the sweet spot for
cartilage formation
A synthetic scafold with ne-tuned physical properties helps cartilage
regrowth at injury sites
Hydrogels offer a promising infrastructure for promoting the repair of damaged cartilage.

2
0
1
4

A
*
S
T
A
R

I
n
s
t
i
t
u
t
e

o
f

B
i
o
e
n
g
i
n
e
e
r
i
n
g

a
n
d

N
a
n
o
t
e
c
h
n
o
l
o
g
y
1. Wang, L.-S., Du, C., Toh, W. S.,
Wan, A. C. A., Gao, S. J. &
Kurisawa, M. Modulation of
chondrocyte functions and
stiffness-dependent cartilage
repair using an injectable
enzymatically crosslinked hydrogel
with tunable mechanical
properties. Biomaterials 35,
22072217 (2014).
Our ndings highlight
the importance
of incorporating
considerations of
hydrogel stifness into
the design of scafolds
intended for cartilage
tissue engineering.
TECHNOLOGIES
Monoclonal antibodies represent
the largest and fastest-growing seg-
ment of international biopharma.
While these therapeutic agents
are a boon for global healthcare,
productivity constraints pose a
serious challenge for manufac-
turers seeking to make sufficient
amounts for therapeutic applica-
tions. Now, A*STAR researchers
have developed a high-capacity
method to purify monoclonal
antibodies that uses magnetic
nanoparticles and also introduces
new operating conditions
1
.
At present, therapeutic
antibodies are generally purified
by a technique known as protein A
affinity chromatography. The
process yields a high purification
factor typically 99 per cent
but it is slow, thereby creating a
severe productivity bottleneck.
The process is largely hindered
by the low capacity of protein A,
which binds monoclonal anti-
bodies at an average rate of
50 grams per liter of protein A
chromatography media. The
overall purification process
requires unpurified antibodies to
pass through columns packed with
the media in multiple cycles that
can take up to a week.
A research team led by
Pete Gagnon and co-workers from
the A*STAR Bioprocessing Tech-
nology Institute in Singapore have
developed an alternative method
with 1,000 times the capacity of
protein A. The technique involves
the use of polyethylene glycol,
which causes the antibodies to be
deposited on the surface of starch-
coated magnetic nanoparticles
(see image). The particles are
collected in a magnetic field,
undeposited contaminants are
washed away and the purified
antibodies recovered by removing
the polyethylene glycol.
The high capacity of our
nanoparticle method makes it
much faster than column chro-
matography, explains Gagnon.
Instead of the pharmaceutical
industry norm of five to eight
cycles, the new process requires
only one cycle, which takes just
a few hours. This reduction
dramatically increases the produc-
tivity of the new approach over
traditional methods.
The new method also required
the research team to develop new
operating conditions. Polyethylene
glycol has been used for decades to
process antibodies, but it has never
achieved the level of purity needed
for clinical therapeutics. The team
discovered that by elevating the
salt concentration, they could
reduce contaminant levels from
about 250,000 parts per million
to 500: the same level achieved
by protein A. A single follow-on
polishing step using a multimodal
chromatography column further
purified the antibodies to clinical
quality standards.
Gagnon notes the high
potential for adoption of the
new technology by industry. In
addition to solving the long-
standing problem of productivity
for monoclonal antibodies, the
nanoparticle approach can be
applied to many other therapeutic
proteins and also to viral vaccines.
Biopurication:
Small particles, big productivity
Magnetic nanoparticles break the capacity barrier for
antibody purication
A novel separation technique uses starch-coated magnetic nanoparticles (green) and polyethylene glycol to purify monoclonal
antibodies (blue).

2
0
1
3

E
l
s
e
v
i
e
r
1. Gagnon, P., Toh, P. & Lee, J. High
productivity purification of
immunoglobulin G monoclonal
antibodies on starch-coated
magnetic nanoparticles by steric
exclusion of polyethylene glycol.
Journal of Chromatography A 1324,
171180 (2013).
TECHNOLOGIES
Erythropoietin (EPO) a
key regulator of red blood cell
production is widely used
for treating certain cancers and
anemia induced by chronic
kidney disease. However, the
time that EPO resides in the
bloodstream depends on the
extent to which it is modified
by sugar chains containing sialic
acid. This modification, known
as sialylation, prevents liver cells
from taking up and destroying
glycosylated EPO (see image).
EPO fully modified with sugars
containing sialic acid persists in
the bloodstream for about one
hour five times longer than
unsialylated EPO.
Since as much as 80% of
manufactured EPO is discarded
because it is insufficiently
sialylated, optimizing the
sialylation of EPO may reduce
the expense and dosing require-
ments of EPO therapy.
Previously, a team led by
Zhiwei Song of the A*STAR
Bioprocessing Technology Institute
in Singapore increased the sialyla-
tion level of EPO in mammalian
cells deficient in the enzyme
N-acetylglucosaminyltransferase
(GnT I) by introducing a func-
tional GnT I gene to the cell line.
EPO sialylation in the genetically
modified cells was almost 25%
more than that obtained from
wild-type cells when cultured
under laboratory conditions.
The same group has now
improved on the capacity of that
line by using a process called
methotrexate amplification to
boost yields of sialylated EPO
by an additional 2.5-fold
1
.
Moreover, the increased yields can
be achieved under industrially
relevant conditions.
The process involves inacti-
vating the gene that normally
makes dihydrofolate reductase
(DHFR) in the GnT I-deficient
cell line, and then expressing a
segment of foreign DNA that
encodes EPO, DHFR and a
foreign GnT I enzyme. Because
methotrexate inhibits DHFR,
which is essential for cell survival,
treatment of this modified cell line
with increasing concentrations
of methotrexate selects cells that
have multiple copies of the DHFR
gene on the foreign DNA segment.
As these cells also have multiple
copies of the adjacent EPO and
GnT I genes, the process also boosts
levels of fully sialylated EPO.
The researchers tested the sialic
acid content of the EPO produced
from the best line and found that
it was over 60% higher than the
amount produced by an existing
industrial line. Although the yields
obtained are still not high enough
for manufacturing industrially
relevant levels of EPO, Song is
confident that additional rounds
of methotrexate amplification and
modification of the medium and
other culture conditions could
further increase yields. Optimal
glycosylation is an important
consideration for the regulatory
use of many protein drugs, he
adds. So our strategy might find
applications beyond EPO alone.
Biotechnology:
Sugarcoating a protein drug
Mammalian cells have been engineered to produce longer-lived versions of
the therapeutic protein erythropoietin
Predicted structure of glycosylated human erythropoietin (EPO), with the sialylated sugar
chains shown in purple and pink.
I
m
a
g
e

c
o
u
r
t
e
s
y

o
f

t
h
e

U
n
i
v
e
r
s
i
t
y

o
f

G
e
o
r
g
i
a
/
T
h
o
m
a
s

W
o
o
d
s
1. Goh, J. S. Y., Liu, Y., Liu, H., Chan, K. F.,
Wan, C. et al. Highly sialylated
recombinant human erythropoietin
production in large-scale perfusion
bioreactor utilizing CHO-gmt4
(JW152) with restored GnT I
function. Biotechnology Journal9,
100109 (2014).
Optimal glycosylation
is an important
consideration for the
regulatory use of many
protein drugs.
TECHNOLOGIES
Despite decades of research,
scientists have yet to create an
artificial neural network capable
of rivaling the speed and accuracy
of the human visual cortex. Now,
Haizhou Li and Huajin Tang at
the A*STAR Institute for Info-
comm Research and co-workers
in Singapore propose using a
spiking neural network (SNN) to
solve real-world pattern recogni-
tion problems
1
. Artificial neural
networks capable of such pattern
recognition could have broad
applications in biometrics, data
mining and image analysis.
Humans are remarkably good
at deciphering handwritten text
and spotting familiar faces in a
crowd. This ability stems from
the visual cortex a dedicated
area at the rear of the brain that is
used to recognize patterns, such
as letters, numbers and facial
features. This area contains a
complex network of neurons that
work in parallel to encode visual
information, learn spatiotemporal
patterns and classify objects
based on prior knowledge or
statistical information extracted
from patterns.
Like the human visual cortex,
SNNs encode visual information
in the form of spikes by firing
electrical pulses down their
neurons. The researchers showed
that an SNN employing suit-
able learning algorithms could
recognize handwritten numbers
from the Mixed National Institute
of Standards and Technology
(MNIST) database with a
performance comparable to that
of support vector machines the
current benchmark for pattern
recognition methods.
Their SNN has a feedforward
architecture and consists of three
types of neurons: encoding,
learning and readout neurons.
Although the learning neurons are
fully capable of discriminating pat-
terns in an unsupervised manner,
the researchers sped things up by
incorporating supervised learning
algorithms in the computation so
that the learning neurons could
respond to changes faster.
The researchers tested the
performance of the SNN by
challenging it with images from
the MNIST, which contains
60,000 training images and
10,000 testing images of
handwritten numbers (ranging
from zero to nine). After
several training iterations, the
SNN could recognize all the
numbers in the database. The
accuracy of the SNN was high
(around 94 per cent) for training
images and moderate (around
79 per cent) for testing images.
Compared with support vector
machines, the encoding and
learning processes of the SNN
were fast for training images and
moderately fast for testing images.
"We utilized biologically plau-
sible mechanisms to build a cognitive
system that is capable of effective
and efficient cognitive computa-
tions, says Tang. Together with
other related works, this paper paves
the way for constructing a general
structure of spiking neural systems
for cognitive computation.
Articial intelligence:
Quick to recognize
Neural networks that function like the human visual cortex may help realize
faster, more reliable pattern recognition
Artificial neural networks that can more closely mimic the brains ability to recognize pat-
terns potentially have broad applications in biometrics, data mining and image analysis.

j
a
n
u
l
l
a
/
i
S
t
o
c
k
/
T
h
i
n
k
s
t
o
c
k
1. Yu, Q., Tang, H., Tan, K.C. & Li, H.
Rapid feedforward computation
by temporal encoding and
learning with spiking neurons. IEEE
Transactions on Neural Networks
and Learning Systems 24,
15391552 (2013).
Together with other
related works, this
paper paves the way for
constructing a general
structure of spiking
neural systems for
cognitive computation.
TECHNOLOGIES
Many proteins undergo an
assembly-line-style process of
glycosylation as they travel from
a cellular structure called the
endoplasmic reticulum (ER)
to the Golgi apparatus and on
through its various compartments,
after which they are released.
Disruptions in this process can
contribute to a variety of diseases.
Now researchers from A*STAR
have identified a regulatory
mechanism that prevents the
production of glycosylated
proteins that potentially promote
cancerous growth
1
.
The glycosylation process
executed by various enzymes
in a stepwise fashion attaches
complex sugar molecules onto
proteins, which can fundamen-
tally alter the destination and
function of these proteins in
the cell. Many tumors produce
proteins that are known as
Tn antigens: these proteins have
been decorated with a single sugar
molecule by enzymes from the
N-Acetylgalactosaminyltrans-
ferase (GalNAc-T) family.
GalNAc-T enzymes normally
reside within the Golgi, but
Frederic Bard and colleagues
Molecular and Cell Biology in
Singapore previously showed
that these enzymes relocate to
the ER in certain tumors
2
. Here
they target resident proteins for
modification and thus contribute
to Tn production.
To better understand why this
happens, Bards team screened
more than 900 regulatory
proteins that affect Tn levels.
Remarkably, their experiments
uncovered a dozen different
proteins that appear to ensure
that GalNAc-T enzymes are
retained within the Golgi
instead of being delivered to the
ER. We were surprised by the
number of regulators that we
could identify, says Bard. One of
these, a signaling protein called
extracellular signal regulated
kinase 8 (ERK8), appeared to
be particularly important for
restricting Tn accumulation to
the Golgi.
When the researchers inhibited
production of ERK8, they
observed a clear redistribution of
GalNAc-Ts from the Golgi to the
ER, indicating that this protein
normally puts the brakes on such
backward traffic (see image).
After knocking down ERK8,
the relocation pathway responds
like a compressed spring that has
just been released, says Bard.
The subsequent production of
Tn antigens stimulates the active
cell migration observed in aggres-
sive cancers, and ERK8 similarly
appears to act as a critical check
against this behavior in cultured
cancer cells.
The researchers also
noted unusually low ERK8
levels in tumor tissue taken
from breast and lung cancer
patients. Bards team is now
working with animal models
to determine the importance
of this protein as a safeguard
against cancerous growth. We
are building a mouse model
where we can experimentally
remove ERK8 and observe the
effects in healthy and cancer
tissues, says Bard. We predict
that loss of ERK8 will promote
tumor invasiveness.
1. Chia, J., Tham, K. M., Gill, D. J.,
Bard-Chapeau, E. A. & Bard, F. A.
ERK8 is a negative regulator of
O-GalNAc glycosylation and cell
migration. eLife 3, e01828 (2014).
2. Gill, D. J., Tham, K. M., Chia, J.,
Wang, S. C., Steentoft, C. et al.
Initiation of GalNAc-type
O-glycosylation in the endoplasmic
reticulum promotes cancer cell
invasiveness. Proceedings of the
National Academy of Sciences USA
110, E3152E3161 (2013).
Cancer biology:
Exercising restraint to stall
tumor growth
Regulatory mechanisms that help trap an enzyme may prevent the
production of cancer-promoting molecules

R
e
p
r
o
d
u
c
e
d

f
r
o
m

R
e
f
.
1

a
n
d

l
i
c
e
n
s
e
d

u
n
d
e
r

C
C

B
Y

3
.0

(
c
r
e
a
t
i
v
e
c
o
m
m
o
n
s
.o
r
g
/
l
i
c
e
n
s
e
s
/
b
y
/
3
.0
/
l
e
g
a
l
c
o
d
e
)

2
0
1
4

J
.
C
h
i
a

e
t

a
l
.
10 m
10 m
The location of GalNAc-T enzymes (red) and the endoplasmic reticulum (ER; green) in
normal cells (top). When the signaling protein ERK8 is inhibited, GalNAc-Ts move from the
Golgi to the ER (yellow staining; bottom).
We predict that loss
of ERK8 will promote
tumor invasiveness.
Research Highlights
ENGINEERING &
NANOTECHNOLOGY
200 nm
ENGINEERING &
NANOTECHNOLOGY
Current computer technology uses
electronics, but a new laser design based
on a thin-layered silicon chip may help
increase data processing capabilities.

O
l
g
a

M
i
l
t
s
o
v
a
/
H
e
m
e
r
a
/
T
h
i
n
k
s
t
o
c
k
Photonics:
Silicon helps light go through the
right channels
When it comes to data trans-
mission, light is superior to
electronics. An ability to transmit
data in parallel by utilizing
multiple light wavelengths allows
optical fibers to carry more
information than electrical cables.
Computers are currently based on
electronics, but they would benefit
from employing optical signals.
However, for this to become a
reality, it needs to be implemented
on a small scale and result in low
power consumption.
Now, Vivek Krishnamurthy
from the A*STAR Data Storage
Institute in Singapore and his
colleagues have designed a laser
on a microelectronic chip that has
a lower power consumption and a
higher efficiency
1
.
By developing lasers on
silicon, we can combine the
electronic data processing capa-
bility of the microelectronic chip
with the high energy efficiency
of optical communications
over distances ranging from a
few micrometers within a chip
to hundreds of meters in data
centers, says Krishnamurthy.
The processing speed of the
microelectronic chip is limited
by its power consumption; most
of the power is consumed by
the connecting electrical wires
and links. Optical links, on the
other hand, consume practically
no energy but are limited by the
power consumption of the light
source, which is often a laser.
For optical links to be feasible
on a small scale, the electrical
power consumption of lasers must
be reduced, yet still be able to
generate sufficient optical energy
for transmission.
Lasers cannot be made from
silicon as it is a poor light emitter.
Instead, lasers are fabricated by
bonding an active material based
on indium phosphide a good
light emitter to a thin silicon
film. However, because silicon is
better for carrying optical signals,
the light from the laser needs to
be routed through the silicon chip
via optical channels. This requires
fabricating optical channels in
silicon outside the laser region.
Generating light efficiently in
the active medium and efficiently
routing it via the silicon layer
simultaneously reduces the
electrical current required and
increases the power generated.
Calculations show that this
silicon-based design will have
a three to four times higher
light generation efficiency than
competing schemes.
This high efficiency makes the
silicon-based laser design prom-
ising for making optical chips,
which, says Krishnamurthy, is
the next step for the project team.
We have begun the experimental
demonstration of the laser, he
says. Our plan is to integrate this
laser onto our silicon platform
and develop a fully functional
photonic system for applications,
for example, in data communica-
tions and storage.
Improved design of lasers on optoelectronic chips will advance
optical communications
1. Krishnamurthy, V., Wang, Q.,
Pu, J., Loh, T.-H. & Ho, S. T. Optical
design of distributed feedback
lasers-on-thin-film-silicon. IEEE
Photonics Technology Letters 25,
944947 (2013).
ENGINEERING &
NANOTECHNOLOGY
1. Lai, S. C., Yao, K. & Chen, Y. F.
A battery-less photo-detector
enabled with simultaneous
ferroelectric sensing and
energy harnessing mechanism.
Applied Physics Letters 103,
092903 (2013).
Light detectors are used extensively
in daily life as brightness sensors
and as receivers for remote control
devices in electrical gadgets, for
example. However, operating these
detectors requires electrical energy,
which limits their versatility.
Now, Kui Yao and col-
leagues from the A*STAR
and Engineering in Singapore have
developed a photodetector that
can harvest just small quantities of
detected light to generate enough
energy to power a sensing signal
transmission through a radio-
frequency transmitter
1
.
While the energy contained in
a beam of light can be converted
into electricity, this energy is not
usually sufficient to continuously
power an electrical circuit. Even the
use of batteries to power a circuit is
impractical in many circumstances,
explains Yao. Use of photosensors
may take place under extremely
harsh conditions intolerable to
batteries, or involve environmental
monitoring network systems
where it may be too expensive or
unrealistic to maintain batteries for
each sensor.
Operating an electrical circuit
under low-power circumstances
requires a buildup of energy,
which must be generated by the
photodetector. However, commonly
used photodetector materials, which
are based on semiconductors, lose
too much energy for this to occur.
Conventional photodetectors cant
accumulate the minute photovoltaic
energy and then harness it to drive
a load in a sustainable manner,
explains Yao.
To overcome such energy losses,
Yao and colleagues developed
photodetectors made from
ferroelectric compounds. These
insulating materials can separate
electrical charges as well as store
them with low losses. Ferroelectric
detectors can also generate a larger
electrical voltage than semiconduc-
tors, making it easier for them to
power other electrical components.
The researchers connected their
ferroelectric detector to a specially
designed electrical circuit, which is
mechanically opened and closed by a
switch in the form of a piezoelectric
cantilever. Any generated electricity
is temporarily stored in the ferroelec-
tric detector and a capacitor. Once
the electrical charge of the capacitor
is sufficiently high, the cantilever
changes its shape and closes the
electrical circuit. This activates a
commercial radio transmitter.
So far, the teams main challenge
in developing the device has been
to minimize electrical losses.
Remarkably, Yao and his team
have shown that almost 70 per cent
of the accumulated electrical
charge can be retrieved from the
capacitor even ten minutes after
the light source has been switched
off. This advantage provides the
teams device with the potential for
use in a wide range of applications,
such as wireless optical sensors and
monitoring networks.
Energy:
Self-powered wireless
light detectors
A low-power photodetection system can harness enough energy to power
an autonomous sensor and monitoring network

G
l
i
d
a
4
6
/
i
S
t
o
c
k
/
T
h
i
n
k
s
t
o
c
k
Ferroelectric compounds used in photodetectors can help to minimize electrical losses, facilitating a sensor and monitoring network
that can go off the grid.
ENGINEERING &
NANOTECHNOLOGY
Wireless transmission at microwave
frequencies is important for high-
data-rate transmission applications,
such as mobile phone networks,
satellite links and remote imaging.
Now, Xianshu Luo and colleagues
Microelectronics in Singapore have
investigated different designs of
silicon modulators that enable fast
data conversion from electrical to
optical signals
1
.
A key component in a
microwave photonic network is
the modulator, which converts an
electrical signal into an optical
signal. The performance of the
microwave photonic system relies
on the quality of this conversion,
which is determined by factors
such as loss, noise and signal
distortion, explains Luo. As the
modulator acts a bridge between
optical components and silicon-
based electronics, it should be
fabricated on a silicon chip.
The researchers built their
modulators according to standard
specifications used for semi-
conductor electronics. A typical
modulator consists of two small
channels for light so-called
waveguides etched into a
silicon chip (see image). Light
is fed into a waveguide on the
chip, which then splits into two;
modulation occurs when these
two beams are reunited. If the
light passing through one channel
is delayed slightly compared to
that in the other channel, the sig-
nals from both beams will either
cancel each other out or reinforce
each other. This property is used
to generate the 0 and 1 signals
for digital transmission.
In silicon modulators, light
transmission in one waveguide
is delayed by applying a radio
signal, which results in electrical
charges either being added to
or removed from the material
surrounding the waveguide. This
addition or subtraction of charge
modifies the optical properties
of silicon.
Modulators based on the
addition or removal of electrical
charges have different attributes.
While the initial injection of elec-
trical charge carriers charges
that are free to move is fast in
modulators based on the addition
of charges, the carrier recombina-
tion takes time, which slows down
the overall speed. Modulators that
have electrical carriers removed,
reducing the nonlinear optical
effects, experience less noise in the
modulated signals.
The different characteristics of
the two types of modulators mean
that they are suited to different
applications, and the researchers
experiments are helping to
inform this choice. Both designs
are capable of fast speeds, with
the devices under test having
an operation bandwidth of
about 10 gigahertz, according
to Luo. More recently we
have demonstrated similar modu-
lators with even larger bandwidths
of up to 28 gigahertz, which
means that they can work at even
faster rates of data transmission,
he says.
A typical modulator consists of two optical waveguides etched into a silicon chip.
Photonics:
Enabling next-generation
wireless networks
1. Luo, X., Tu, X., Song, J., Ding, L.,
Fang, Q. et al. Slope efficiency and
spurious-free dynamic range of
silicon Mach-Zehnder modulator
upon carrier depletion and injection
effects. Optics Express 21,
1657016577 (2013).
A modulator that converts an electrical signal into an optical signal could
enable faster wireless data transmission

2
0
1
4

A
*
S
T
A
R

I
n
s
t
i
t
u
t
e

o
f

M
i
c
r
o
e
l
e
c
t
r
o
n
i
c
s
More recently we have
demonstrated similar
modulators with even
larger bandwidths of up
to 28 gigahertz, which
means that they can
work at even faster rates
of data transmission.
ENGINEERING &
NANOTECHNOLOGY
In the eighteenth century,
scientists faced a conundrum: is
light a wave or a particle? One of
the strongest pieces of evidence
to support the wave view the
landmark double-slit experiment
was reported in 1804 by the
scientist Thomas Young. Young
passed coherent light through two
closely spaced slits and observed a
set of interference fringes, a result
that occurs with wave phenomena
like sound or water. This observa-
tion became the basis for the
modern wave theory of light.
Two hundred years later,
Arseniy Kuznetsov and co-
workers from the A*STAR
Data Storage Institute, together
with collaborators in Australia,
Singapore, the United Kingdom
and Russia, have performed an
experiment analogous to Youngs
experiments but using nanoscale
objects
1
. The team studied the
light scattering in the visible and
near-infrared wavelength regions
from a cluster of two or three
closely spaced gold plasmonic
nanoparticles. They observed
interference and resonance effects
that resemble those seen in
Youngs experiments.
In particular, while studying
a trimer system consisting of
three discrete metallic nanodisks
of about 145 nanometers in
diameter and 60 nanometers
thick, the team found evidence
for the presence of near-field,
subwavelength-sized optical
vortices and the circulation
of electromagnetic energy
(see image). This finding is very
similar to what occurs to the
energy f low pattern in a Young-
type experiment performed with
three slits.
One of the key issues in
nanoplasmonics is the interaction
between metallic nanoparticles
at the nanoscale. Even if
the separation between two
or multiple non-periodically
arranged nanoparticles is of
the order of wavelength, their
interaction can be strong enough
to change their scattering
and absorption properties,
notes Kuznetsov. This can be
explained by the peculiarities
of the Poynting vector (energy)
f low around the nanoparticles
and formation of optical vortices,
which produce a pattern of
field lines similar to Youngs
classic experiment.
The teams findings, says
Kuznetsov, not only expand our
fundamental understanding of
how light interacts with nanoclus-
ters of metallic particles, but have
both theoretical and practical
applications. They may also
prove useful for applications such
as improved solar cells and plas-
monic biosensors. However, their
most remarkable application, he
suggests, may be in the emerging
area of nanoantennas.
In the future, the team is
aiming to study the resonance
properties and interactions
of nanoparticles made from
nonmetallic materials. In
particular, they plan to
investigate high-refractive index
dielectric materials such as
silicon, which, unlike metallic
particles, do not suffer from high
optical losses.
1. Rahmani, M., Miroshnichenko, A. E.,
Lei, D. Y., Lukyanchuk, B.,
Tribelsky, M. I. et al. Beyond the
hybridization effects in plasmonic
nanoclusters: Diffraction-induced
enhanced absorption and scattering.
Small 10, 576583 (2013).
Plasmonics:
A modern twist on Youngs slits
A landmark experiment on wave interference from the early 1800s is
revisited using gold nanoparticles

2
0
1
4

A
*
S
T
A
R

D
a
t
a

S
t
o
r
a
g
e

I
n
s
t
i
t
u
t
e
Example of the energy flow and optical vortices found around closely spaced gold
nanoparticles. The effects resemble the field lines seen in Youngs slit experiments.
This will get us closer
to our ultimate goal of
using solar illumination
as an abundant source of
renewable energy.
ENGINEERING &
NANOTECHNOLOGY
Electronics:
More speed,
less interference
A semi-analytic model can compute electromagnetic interference on an
electronic circuit board ten times faster than existing commercial software
The capacity to effectively model
electromagnetic interference on
electronic circuit boards in the product
design phase will save time and money
for the electronics industry.

V
v
o
e
V
a
l
e
/
i
S
t
o
c
k
/
T
h
i
n
k
s
t
o
c
k
As electronic components on
electronic circuit boards continue to
shrink, problems of electromagnetic
compatibility are arising. Such
problems include unwanted noise
effects due to electromagnetic
interference and susceptibility.
Electromagnetic interference is a
critical problem for the electronics
industry, explains Xian-Ke Gao
High Performance Computing in
Singapore. Engineers are keen to
understand how the electronic cir-
cuits react. However, it is difficult to
measure such effects experimentally,
because disassembling the device
would affect the physical testing.
To address this problem, the
electronics industry has developed
a suite of computer modeling
tools, but these are cumbersome
and require a lot of computing
power. Now, Gao and colleagues
have developed a computer model
that is able to solve such problems
more than ten times faster than
existing models
1
.
Fairly coarse models are
typically used to model electro-
magnetic interference effects
on electronic circuit boards
(see image). To do this, the device
is divided into a grid of small
cubes, and the electromagnetic
fields to and from each cube
are modeled individually.
This approach requires a lot of
computing power, especially if the
grid size is small, but it has the
advantage that it is flexible and can
be adapted to various geometries.
Except for interference effects,
the same computer models can be
applied to calculate electromag-
netic fields for a range of electrical
devices other than circuit boards.
A more targeted and efficient
approach is required to measure
interference effects. Researchers
use mathematical equations to
describe the electrical currents in
a conducting wire. The physics of
these transmission-line equations
are well understood and, once
adapted to the unique properties of
circuit boards, are far easier to solve
by a computer algorithm than the
other, coarser modeling.
The first tests of the software
package developed by the A*STAR
researchers, which is based on
the transmission-line equations,
reliably solved a number of standard
problems for electronic circuits.
Compared to commercial models,
the new software achieved very
good agreement, especially for the
main region of interest frequen-
cies below one gigahertz.
Speed, however, is the key
advantage of using the software.
Whereas commercial software
requires more than two hours of
computing on a regular laptop,
the A*STAR software package
needed less than ten minutes for
the same task, explains Gao. Our
computational problem-solving
kit can shorten electromagnetic
interference trouble-shooting in the
product design phase and therefore
translates into time and cost savings
for the industry.
1. Gao, X.-K., Zhao, H., Li, E.-P. &
Hoefer, W. J. R. Radiated
electromagnetic immunity analysis
of flex cable with ground plane using
transmission line equations. IEEE
Transactions on Electromagnetic
Compatibility 55, 875882 (2013).
ENGINEERING &
NANOTECHNOLOGY
Labeling biomolecules with light-
emitting nanoparticles is a powerful
technique for observing cell move-
ment and signaling under realistic,
in vivo conditions. The small size
of these probes, however, often
limits their optical capabilities. In
particular, many nanoparticles have
trouble producing high-energy light
with wavelengths in the violet to
ultraviolet range, which can trigger
critical biological reactions.
Now, an international team led
by Xiaogang Liu from the A*STAR
and Engineering and the National
University of Singapore has
discovered a novel class of rare-
earth nanocrystals that preserve
excited energy inside their atomic
framework, resulting in unusually
intense violet emissions
1
.
Nanocrystals selectively infused,
or doped, with rare-earth ions
have attracted the attention of
researchers, because of their low
toxicity and ability to convert low-
energy laser light into violet-colored
luminescence emissions a process
known as photon upconversion.
Efforts to improve the intensity of
these emissions have focused on
ytterbium (Yb) rare-earth dopants,
as they are easily excitable with
standard lasers. Unfortunately,
elevated amounts of Yb dopants can
rapidly diminish, or quench, the
generated light.
This quenching probably arises
from the long-range migration
of laser-excited energy states
from Yb and toward defects in
the nanocrystal. Most rare-earth
nanocrystals have relatively
uniform dopant distributions, but
Liu and co-workers considered
that a different crystal arrange-
ment clustering dopants into
multi-atom arrays separated by
large distances could produce
localized excited states that do not
undergo migratory quenching.
The team screened numerous
nanocrystals with different
symmetries before discovering a
material that met their criteria: a
potassium fluoride crystal doped
with Yb and europium rare earths
(KYb
2
F
7
:Eu). Experiments revealed
that the isolated Yb energy clusters
inside this pill-shaped nanocrystal
(see image) enabled substantially
higher dopant concentrations than
usual Yb accounted for up to
98 per cent of the crystals mass
and helped initiate multiphoton
upconversion that yielded violet
light with an intensity eight times
higher than previously seen.
The researchers then explored
the biological applications of
their nanocrystals by using them
to detect alkaline phosphatases,
enzymes that frequently indicate
bone and liver diseases. When the
team brought the nanocrystals
close to an alkaline phosphate-
catalyzed reaction, they saw the
violet emissions diminish in
direct proportion to a chemical
indicator produced by the enzyme.
This approach enables swift and
sensitive detection of this critical
biomolecule at microscale concen-
tration levels.
We believe that the funda-
mental aspects of these findings
that crystal structures can
greatly influence luminescence
properties could allow upconver-
sion nanocrystals to eventually
outperform conventional fluores-
cent biomarkers, says Liu.
Nanocrystals:
The promise of purple for
enhanced bioimaging
Newly detected energy-clustering structures inside rare-earth nanoparticles generate
intense violet light, which is ideal for studying photon-induced transformations
200 nm
A novel type of pill-shaped nanocrystal emits the correct light frequencies for triggering and detecting many biological reactions.
R
e
p
r
o
d
u
c
e
d
,
w
i
t
h

p
e
r
m
i
s
s
i
o
n
,
f
r
o
m

R
e
f
.
1

2
0
1
4

J
.
W
a
n
g

e
t

a
l
.
1. Wang, J., Deng, R., MacDonald, M. A.,
Chen, B., Yuan, J. et al. Enhancing
multiphoton upconversion through
energy clustering at sublattice
level. Nature Materials 13,
157162 (2014).
ENGINEERING &
NANOTECHNOLOGY
Modulators are key components
within optical fiber networks and
serve to transfer information from
an electrical current to a signal
suitable for optical fibers. They
function by turning a light beam
on and off quickly and the faster
they can do this, the more data
that can be transmitted.
An increase in data traffic
creates a need for a reduction in
the cost and size of optical com-
ponents. An improved low-loss
design for modulators, suitable
for silicon computer chips, has
been developed by Soon Thor Lim
and colleagues from the A*STAR
Computing, the A*STAR Institute
of Microelectronics and collabora-
tors from Fujikura Ltd
1
.
Existing optical modulators are
based on lithium niobate, a material
that is expensive and unsuitable for
silicon chips. While silicon offers an
inexpensive alternative, it can only
be used with the addition of other
elements that can create positive and
negative movable electrical charges.
Modulation requires the movable
charges to be channeled in and out
of the device by an alternating elec-
trical voltage, which controls both
the speed of light through the chip
and the data rate. Light that passes
through this region and crosses light
that passes through a neutral silicon
region creates interference effects in
the optical beam that switches the
light on and off.
Previous modulator designs
contained charged regions that
were relatively large, with the
drawback that they increased light
absorption in the chip. However,
in the teams proposed design,
this area is reduced so that less of
the laser beam passes through the
charged region (see image).
After computer modeling the
performance of the modulator,
the team fabricated their device
on a silicon chip that has light
channels only 220 nanometers
high and 550 nanometers wide.
Compared to designs with large
charged areas, these modulators
reduced optical losses by up to
28 per cent and operated at faster
speeds of 10 gigabits per second.
Our device has a speed
and optical losses comparable
to existing technology such as
lithium niobate, says Lim. One
reason for this high performance
is because we used highly accurate
computer codes developed
in-house.
Successfully demonstrating
the device also highlights how
modeling software can reduce
the required number of experi-
ments, Lim adds. Simulation
and analysis helps to visualize
the physical behavior of these
cutting-edge optical devices. This
can identify potential problems
and circumvent the need for
costly multiple design iterations
ultimately accelerating the speed
to market.
Photonics:
Clear predictions
Computer modeling is assisting the design of optical modulators with low
losses to improve optical communications
The two plots show the distribution of electrical charges in different modulator designs.
Compared to a previous design (top), the modulator improved by computer modeling
(bottom) contains fewer free electrical charges, as depicted by lighter shades of blue
(positive charge) and yellow (negative charge).

2
0
1
4

A
*
S
T
A
R

I
n
s
t
i
t
u
t
e

o
f

H
i
g
h

P
e
r
f
o
r
m
a
n
c
e

C
o
m
p
u
t
i
n
g
1. Goi, K., Ogawa, K., Tan, Y. T.,
Dixit, V., Lim, S. T. et al. Silicon
MachZehnder modulator using
low-loss phase shifter with bottom
PN junction formed by restricted-
depth doping. IEICE Electronics
Express 10, 20130552 (2013).
ENGINEERING &
NANOTECHNOLOGY
Grid computing is a powerful form
of distributed computing wherein
a network of loosely coupled and
geographically separated computers,
typically of different computational
powers, work together to perform
data-intensive calculations. The
technology uses numerical simula-
tions to help investigate a variety
of challenging scientific problems,
including the subatomic world
revealed by particle accelerators like
the Large Hadron Collider.
The current trend in the
industry is to build grid-computing
systems that can run not just one
but multiple large-scale, numerical
simulations. While most systems
can guarantee good performance,
this is usually accompanied by
significant cost and large storage
requirements. To optimize the
costperformance relationship of
large-scale grid-computing systems,
scientists must overcome several
issues one of which is the effi-
cient allocation of computational
resources, known as scheduling.
Rubing Duan and Xiaorong Li
at the A*STAR Institute of High
Performance Computing in
Singapore and co-workers have now
developed a scheme to address the
scheduling problem in two large-
scale applications: the ASTRO pro-
gram from the field of cosmology,
which simulates the movements
and interactions of galaxy clusters,
and the WIEK2k program from the
field of theoretical chemistry, which
calculates the electronic structure of
solids
1
. The researchers new scheme
relies on three game-theory-based
scheduling algorithms: one to
minimize the execution time; one to
reduce the economic cost; and one
to limit the storage requirements.
The researchers performed
calculations wherein they stopped
the competition for resources when
the iteration reached the upper limit
of optimization. They compared
their simulation results with those
from related algorithms namely,
Minimum Execution Time,
Minimum Completion Time,
Opportunistic Load Balancing,
Max-min, Min-min and Suf-
ferage. The new approach showed
improvements in terms of speed,
cost, scheduling results and fairness.
Furthermore, the researchers found
that the execution time improved
as the scale of the experiment
increased. In one case, their
approach delivered results within
0.3 seconds while other algorithms
needed several hours.
Nonetheless, the researchers
note that their algorithms may not
be suited for applications that are
highly heterogeneous.
Prior to this study, only a
handful of work had considered the
optimization of multiple metrics.
However, Duan, Li and colleagues
were able to present a model
approach to reducing the execution
time, economic cost and storage
requirements in grid-computing
systems, especially when used in
large-scale applications.
Our game-theory-based
scheduling algorithms possess great
potential for large-scale applica-
tions, says Duan. We are looking
into how the algorithms adapt to
other metrics, such as memory,
security, resource availability,
network bandwidth and multiple
virtual organizations.
Computer science:
Right on schedule
Scheduling algorithms based on game theory can make better use of
computational resources to reduce the costs of grid computing
A new scheme developed for the efficient allocation of computational resources relies
on three game-theory-based scheduling algorithms that help minimize execution time,
reduce cost and limit storage requirements.

a
g
s
a
n
d
r
e
w
/
i
S
t
o
c
k
/
T
h
i
n
k
s
t
o
c
k
1. Duan, R., Prodan, R. & Li, X.
A sequential cooperative game
theoretic approach to scheduling
multiple large-scale applications in
grids. Future Generation Computer
Systems 30, 2743 (2014).
We are looking into how
the algorithms adapt
to other metrics, such
as memory, security,
resource availability,
network bandwidth
and multiple virtual
organizations.
ENGINEERING &
NANOTECHNOLOGY
Electric vehicles are becoming
more popular due to their
environmental credentials and
relatively low running costs.
However, most existing electric
vehicles need to be recharged
every 100 to 150 kilometers,
with each recharge potentially
exposing information related to a
customers payment and location.
Now, researchers at A*STAR
have described a new system
that would allow quick and
easy money transfers at electric
vehicle charging stations, without
jeopardizing customer privacy.
Cybersecurity is an important
factor for payment systems, but
it is often ignored by users or
administrators until the system is
being attacked, says researcher
Joseph Liu from the A*STAR
Institute for Infocomm Research
in Singapore
1
. No one should
have their daily habits or behavior
traced without their consent.
The recharging of electric vehi-
cles presents unique challenges
for privacy, not least because
some cars with solar panels are
able to sell electricity back to the
grid, meaning payments f low in
both directions. Without tight
security, payment companies or
hackers could monitor where and
when cars are charged, gaining
insight into peoples lifestyles that
could be exploited for targeted
spam marketing.
Some popular electronic
payment systems like credit cards
do not provide any privacy, while
other systems like prepaid cash
cards may not be suitable for
large payments, or are not insured
against card loss, says Liu.
Cash is anonymous, but requires
expensive machines to keep cash
stores secure from thieves.
The new system developed by
Liu and co-workers is based on
an in-car unit that resembles a
smartphone or tablet and, along
with a range of security benefits,
allows two-way anonymous
payments for recharging. Users
can instantly shut down their
accounts and retrieve unused
credit. Also, if their car is stolen
they can revoke the location
privacy to help police trace the
car. In the event of a dispute
between a user and a supplier,
either party can submit the
claims to an independent judging
authority for investigation.
The researchers tested their
system by simulating three dif-
ferent types of attack: a hacker
trying to track the transactions
of an honest user, a user trying to
underpay for services, and a sup-
plier trying to slander an honest
user. The system proved robust
against all three attacks.
The team has now imple-
mented a prototype of their secure
charging system. They will install
the tamper-proof in-car units on
a f leet of 100 new electric vehicles
that will arrive in Singapore later
this year, thanks to collaboration
with the Chinese carmaker
BYD Auto.
Electric vehicles:
Recharging in private
An electronic payment system developed at A*STAR will protect the privacy
of customers recharging their electric vehicles
Electric vehicles require frequent
recharging, which presents challenges
for privacy and data protection.

M
i
h
a
j
l
o

M
a
r
i
c
i
c
/
i
S
t
o
c
k
/
T
h
i
n
k
s
t
o
c
k
1. Au, M. H., Liu, J. K., Fang, J., Jiang, Z. L.,
Susilo, W., & Zhou, J. A new
payment system for enhancing
location privacy of electric vehicles.
IEEE Transactions on Vehicular
Technology 63, 318 (2014).
ENGINEERING &
NANOTECHNOLOGY
A visual saliency technique that
can detect and extract relevant
information from both still
and moving images has many
applications for computer image
processing. Such a technique
can be used to detect motion,
distinguish different objects
and improve the quality of
specific parts of an image through
selective compression.
Shijian Lu and colleagues
from the A*STAR Institute for
Infocomm Research in Singapore
have developed a robust and
efficient method for capturing
such salient information from
images and movies. They found
that the key lies in the distribution
of brightness and color between
pairs of pixels
1
.
Digital images are encoded
as pixels, or points in an image.
To detect an object (for example,
a person standing in the fore-
ground), brightness variations
between neighboring pixels
could be compared. However,
considering just individual
pixels can be deceiving, as the
context is important when
seeking to distinguish between
important details and unimportant
background information.
The technique developed by the
A*STAR researchers hence involves
counting the pixels in an image
based on their color and then plot-
ting the distribution. This makes
not only the distribution of colors
apparent, but also the frequency at
which pairs or neighboring pixels
appear. A low frequency of pixel
pairs with a certain difference in
color indicates a region of high
interest, as it denotes clear bounda-
ries between objects. In this way,
the salient features can be easily
identified; for example, not only
large areas of contrast in a photo-
graph, such as a yellow school bus
in front of a neutral background,
but also contrasts in smaller areas,
such as a person wearing a safety
vest (see image).
Our model has great potential
for predicting the point in an
image that will attract the human
eye, comments Lu. Apart from
generic object detection, it can be
applied to tasks such as guiding
robots or to the smart design of
web pages and advertisements.
The next step for the
researchers will be to apply this
scheme to detecting motion in
videos, which follows similar rules
as identifying relevant information
in still photographs. Moreover, Lu
says that their algorithm enables
more complex approaches to
image analysis.
An example is the development
of computational modeling of the
top-down approach of humans
looking at a scene, says Lu. Com-
bining our bottom-up modeling
algorithm with a top-down visual
search could solve many chal-
lenging computer vision problems
such as anomaly detection or
target search in a more robust,
efficient and cognitive manner.
Image processing:
Focusing on whats relevant
Analyzing pixel correlations in photographs outperforms current methods of
extracting relevant information for image analysis
Saliency map (right) of an image (left), which illustrates how a computer model is able to identify salient information such as the
high-visibility vest from an image based on statistical analysis.

2
0
1
4

A
*
S
T
A
R

I
n
s
t
i
t
u
t
e

f
o
r

I
n
f
o
c
o
m
m

R
e
s
e
a
r
c
h
1. Lu, S., Tan, C. & Lim, J.-H. Robust and
efficient saliency modeling from
image co-occurrence histograms.
IEEE Transactions on Pattern
Analysis and Machine Intelligence
36, 195201 (2014).
Combining our
bottom-up modeling
algorithm with a
top-down visual search
could solve many
challenging computer
vision problems such
as anomaly detection
or target search in a
more robust, efcient
and cognitive manner.
ENGINEERING &
NANOTECHNOLOGY
Antireflection coatings are familiar
from their use in everyday optical
devices, such as glasses and lenses.
They can increase the amount of
light that passes through optical
instruments by reducing the frac-
tion of light reflected (and hence
lost) at surfaces. Antireflection
coatings have applications beyond
visible light: for instance, in the
infrared and terahertz regimes
they are useful for chemical
sensing and imaging applications,
such as those employed at airport
security checks.
Now, Jing Hua Teng from the
A*STAR Institute of Materials
Research and Engineering and
colleagues from the A*STAR
Institute of Microelectronics and
Osaka University, Japan, have
developed ultrathin antiref lection
coatings for terahertz waves that
can be applied to almost any
surface
1
. Their fabrication is very
straightforward, as it takes only
one step of photolithography,
metal deposition and lift-off,
explains Teng.
Antiref lection coatings are
usually based on interference
effects, which requires them to be
at least as thick as the wavelength
of light. This is practical for
visible light, with wavelengths
in the range of hundreds of
nanometers. However, it is a
serious limitation for infrared or
terahertz radiation, which has
much longer wavelengths of the
order of hundreds of microns.
Moreover, as these coatings are
often functional only over
narrow frequency ranges, they
do not operate across the broad
ranges needed for terahertz
sensing applications.
The research team developed
antiref lection coatings based
on metamaterials, which are
metallic structures that are much
smaller than the wavelength
used. These structures completely
alter the optical properties of a
material in a predetermined way,
enabling the generation of a much
broader range of optical effects
than those that occur naturally.
One application of the unusual
optical effects they produce is
invisibility cloaks.
In the new design for
metamaterial surfaces developed
by the researchers, thin strips
of chromium are fabricated
on a silicon surface to form a
grating (see image). Silicon,
being f lexible, is a typical mate-
rial for terahertz optics. When
terahertz light passes through
the stripes and into the silicon,
its phase is changed in the
same way as for the much thicker
coatings based on interfer-
ence effects; this suppresses
surface ref lection.
These metasurfaces have the
advantage that they can function
across an unprecedentedly wide
frequency range, namely 0.06 to
3 terahertz. The f lexibility of the
coatings for other wavelengths
and applications also enhances
their commercial appeal, com-
ments Teng. The beauty of this
method is that it is very f lexible
and can be easily adapted to other
metals and substrates.
Photonics:
Cutting reflection for the
infrared and beyond
Efcient broadband antireection coatings improve terahertz optics for
applications in sensing and imaging
Terahertz
light
With
coating
Without
coating
Reection
Minimized
reection
Thin strips of chromium form an efficient antireflection coating for terahertz light and
can be applied to a broad range of surfaces.

2
0
1
4

A
*
S
T
A
R

I
n
s
t
i
t
u
t
e

o
f

M
a
t
e
r
i
a
l
s

R
e
s
e
a
r
c
h

a
n
d

E
n
g
i
n
e
e
r
i
n
g
1. Ding, L., Wu, Q. Y. S., Song, J. F.,
Serita, K., Tonouchi, M. &
Teng, J. H. Perfect broadband
terahertz antireflection by deep-
subwavelength, thin, lamellar
metallic gratings. Advanced Optical
Materials 1, 910914 (2013).
Their fabrication is very
straightforward, as it
takes only one step of
photolithography, metal
deposition and lift-of.
ENGINEERING &
NANOTECHNOLOGY
Computer numerically controlled
(CNC) milling machines used
in high-tech industries such as
aeronautical manufacturing
operate continuously to maximize
the production efficiency. However,
this requires careful monitoring
for any tool deterioration. Omid
Geramifard and co-workers from
the Singapore Institute of Manu-
facturing Technology at A*STAR
have now developed an improved
algorithm for diagnosing tool-wear
problems before they occur
1
.
CNC milling machines cut and
shave metal materials into precisely
specified structures. To ensure they
can operate 24 hours a day without
any unnecessary downtime, the
milling tools are carefully monitored
with unobtrusive sensors and
analytical computer models. The
algorithm developed by Gerami-
fards team uses a sophisticated,
multicomponent model that narrows
down tool sensor data to the most
effective group for analysis an
innovation that boosts its predictive
capabilities while maintaining its
computational efficiency.
As it is hard to estimate when
tools will fail by simply using
physical models of actual milling
machines, researchers tend to use
data-driven models that analyze
historical tool-wear patterns.
One such method, known as the
hidden Markov model (HMM),
hypothesizes that a tools condition
depends only on its past behavior, a
simplification known as the Markov
assumption. Hidden states are then
introduced to account for degrada-
tion factors that cannot be inspected
directly. Using probability equations
to relate observable data to the
hidden states, HMMs can calculate
how tools change over time.
Geramifard and co-workers
improved current models by
developing a multimode HMM-
based approach that captures and
analyzes several different tool-wear
parameters. They trained each
mode of the HMM with selected
segments of experimental data, and
then combined the multiple outputs
using weighting schemes. However,
running simultaneous HMMs
requires intense computational
power, which significantly slows the
CNC machine monitoring process.
To overcome this problem, the
researchers devised a windowing
technique that reduces the compu-
tational cost by performing HMM
calculations over a short time
frame selected through a cross-
validation process in the training
phase instead of over the full
observation sequence. When
the team used their multimodal
approach to predict tool wear in a
real CNC milling machine, they
found their approach outperformed
conventional techniques when
appropriate window lengths were
adopted. This breakthrough in
accuracy was realized by removing
unnecessary connections to
old observations.
This model opens the path to
more effective stochastic modeling
of tool wear and degradation, says
Geramifard. With more accurate,
data-driven diagnostics and predic-
tion, more efficient tool usage can
occur for assured quality of the
produced workpiece.
Data mining:
Getting the edge on tool wear
A multimode algorithm helps to get a better handle on tool deterioration
while maintaining a low computational cost
An innovative algorithm can improve the productivity of computer-controlled milling
machines used in high-tech manufacturing.

2
0
1
4

A
*
S
T
A
R

S
i
n
g
a
p
o
r
e

I
n
s
t
i
t
u
t
e

o
f

M
a
n
u
f
a
c
t
u
r
i
n
g

T
e
c
h
n
o
l
o
g
y
1. Geramifard, O., Xu, J.-X., Zhou, J.-H.
& Li, X. Multimodal hidden Markov
model-based approach for tool
wear monitoring. IEEE Transactions
on Industrial Electronics 61,
29002911 (2014).
With more accurate,
data-driven diagnostics
and prediction,
more efcient tool
usage can occur for
assured quality of the
produced workpiece.
ENGINEERING &
NANOTECHNOLOGY
Semiconductor optical devices
are becoming increasingly
commonplace. For example,
light-emitting diodes, as they
become more power efficient, are
rapidly replacing conventional
light bulbs. Lasers too are now
found in every barcode scanner
and compact-disc reader.
When designing these devices,
a crucial consideration is how best
to get the light generated within
the solid material out into the
real world. Chee-Wei Lee at the
A*STAR Data Storage Institute,
Singapore, and international
colleagues have now proposed a
light-extraction scheme that is
capable of transferring over half the
light created by a submicrometer-
scale laser into a waveguide
1
.
Plasmonic lasers are the
smallest lasers created to date
they can even be smaller
than the wavelength of the light
they emit. This counterintuitive
property results from plasmons,
which are hybrid electronlight
particles created by coupling light
with electrons in a metal.
Lee and his team considered
the simplest plasmonic laser: a ring
of a light-emitting semiconductor
coated with a thin silver layer.
Light can travel round and round
inside the ring, which provides
the optical cavity required in
most laser devices. What is more,
this tiny laser can be bonded
onto a silicon substrate to make
it compatible with compact
photonics-on-a-chip technology.
Lee and his team used computer
simulations to demonstrate that
high extraction efficiency is
obtained when a waveguide (a
light-carrying submicrometer-wide
semiconductor strip) is directly
connected to the side of the laser.
The team used a numerical
simulation technique called
finite-difference time-domain
to study the performance of
waveguides of different widths
connected at different points on
the laser. Their models revealed
that the optimal structure is an
asymmetric one. When the extrac-
tion waveguide is displaced from
the center of the ring so that
the waveguide is flush with the
edge of the cavity it produces
a peak out-coupling efficiency
of 56 per cent (see image). Our
scheme, based on directly joining
a waveguide, enhances light
extraction by splitting the plasmon
mode, explains Lee.
Scientists have previously
extracted light from plasmonic
lasers by running a waveguide
extremely close to, but not
touching, the cavity ring. Light
can leak across the gap between the
laser and the waveguide through
an effect called evanescent cou-
pling. But this approach requires
precise control over the gap size
and the optical properties of the
material in the gap. The method
developed by the team, however,
can be implemented using much
simpler device fabrication. We
are now in the process of actually
realizing such a device, says Lee.
Nano-optics:
Getting the most out of
tiny lasers
An of-center waveguide enables light to be efciently extracted from
nanoscale lasers
Computer simulations show that efficient light extraction from a nanoring plasmonic
laser can occur when a waveguide is connected flush with one edge of the device.
M
o
d
i
e
d

f
r
o
m

R
e
f
.
1

a
n
d

l
i
c
e
n
s
e
d

u
n
d
e
r

C
C

B
Y
-
N
C

3
.0

(
c
r
e
a
t
i
v
e
c
o
m
m
o
n
s
.o
r
g
/
l
i
c
e
n
s
e
s
/
b
y
-
n
c
/
3
.0
/
l
e
g
a
l
c
o
d
e
)

2
0
1
4

C
.
L
e
e

e
t

a
l
.
1. Lee, C.-W., Singh, G. & Wang, Q.
Light extraction a practical
consideration for a plasmonic
nano-ring laser. Nanoscale 5,
1083510838 (2013).
Our scheme, based
on directly joining a
waveguide, enhances light
extraction by splitting the
plasmon mode.
ENGINEERING &
NANOTECHNOLOGY
The ability to detect tiny quanti-
ties of molecules is important
for chemical sensing as well as
biological and medical diagnostics.
In particular, some of the most
challenging and advanced applica-
tions involve rare compounds for
which only a few molecules may
be present at a time. The most
promising devices for achieving
ultrahigh-precision detection are
nanoscale sensors, where molecules
are placed in tiny gaps between
small gold plates. But this method
is effective only if the molecules
are positioned accurately within
the gaps. Now, Jinghua Teng from
the A*STAR Institute of Materials
Research and Engineering,
Singapore, and colleagues from the
National University of Singapore
have developed a sensor where
molecules are efficiently guided
and placed into position
1
.
The electronic resonances
occurring in gold nanostructures
are like very powerful antennas,
able to amplify radiation from
small molecules in their vicinity.
This permits even the detection
of single molecules. In order for
the signal to be picked up by the
antennas, however, the molecules
need to be precisely located
within electromagnetic hot spots
(see image). We approached this
challenge and developed a method
to selectively bind the molecules
to the electromagnetic hot spots
in the nanoantenna structure for
maximum effect, explains Teng.
The researchers needed to
prepare the device surface such
that the molecules bind only to
the desired areas between the
gold plates not on them. They
achieved this by depositing a thin
titanium film between the gold
plates. The titanium oxidizes
in air, forming stable titanium
dioxide, which is insulating and
has very different properties to the
gold plates. The researchers then
covered the surface with various
organic solutions that selectively
prevent proteins and other
molecules from binding to the gold
while attracting the molecules of
interest to the titanium pad. In
initial tests, signals from molecules
attached to the titanium in the
hot spot showed a six times higher
sensitivity than those randomly
attached across the device.
The next step will be to
increase the sensor sensitivity to
the ultimate limit, explains Teng.
People have been dreaming
of and working toward single-
molecule sensing. This work is
part of these efforts. It provides
a way to selectively bind biomol-
ecules to the hot spots and proves
that it can enhance the molecular
sensitivity and reduce the amount
of sample required. Further
improvements in device design
will however be required, adds
Teng. Moving forward, we would
like to further push the sensitivity
by optimizing the structure
and try multi-agent sensing in
one chip.
Photonics:
Hot spots for molecules
The accurate placement of molecules into gaps between gold nanoantennas
enables ultrahigh-sensitivity molecular detection
Titanium pads (pink) placed in the hot spots (red) between oval gold plates can be used to sense tiny amounts of molecules (blue).

2
0
1
4

A
*
S
T
A
R

I
n
s
t
i
t
u
t
e

o
f

M
a
t
e
r
i
a
l
s

R
e
s
e
a
r
c
h

a
n
d

E
n
g
i
n
e
e
r
i
n
g
1. Zhang, N., Liu, Y. J., Yang, J.,
Su, X., Deng, J. et al. High sensitivity
molecule detection by plasmonic
nanoantennas with selective
binding at electromagnetic
hotspots. Nanoscale 6,
14161422 (2014).
ENGINEERING &
NANOTECHNOLOGY
Microscopes are conventionally
used to image tiny features. How-
ever, their resolution is inherently
limited by the wavelength of light.
This limitation means that they
can resolve only structures larger
than a few hundred nanometers.
Now, Leonid Krivitsky and
Boris Lukyanchuk at the A*STAR
Data Storage Institute in Singapore
and co-workers have demonstrated
an alternative optical approach
capable of mapping surfaces at
resolutions below 100 nanometers
1
.
Diffraction is the tendency for
all waves, including light, to
spread out when they pass near
an object or through a gap. This
effect means that optical imaging
systems cannot resolve objects
smaller than roughly half the
wavelength of the illuminating
light. Thus, for red light with a
wavelength of about 600 nanom-
eters, the resolution will be
approximately 300 nanometers.
Lukyanchuk and his col-
leagues previously showed that a
micrometer-scale transparent bead
placed on a surface can circum-
vent this so-called diffraction
limit. They demonstrated that
light passing through the bead,
when collected by a conventional
microscope, can create an image
of the surface beneath it with
a resolution of 50 nanometers.
However, generating a complete
two-dimensional map requires
scanning the bead across the
surface not easy to perform in
a controlled way when the sphere
is only 6 micrometers across. We
have now improved this superreso-
lution technique by developing
a method to controllably move
the imaging microspheres,
says Krivitsky.
Krivitsky and his team
accomplished such spatial scan-
ning using a tiny pipette with a
tip just 1 or 2 micrometers wide.
Computer simulations confirmed
that the presence of the pipette
would not adversely affect the
superresolution capability of
the microspheres. To fasten the
pipette to the bead, they sucked
the air out from within its
cavity (see image).
The team then connected
the other end of the pipette to a
mechanical stage, which could
move in steps as small as
20 nanometers. Importantly, the
vacuum inside the pipette created
a bond tight enough to ensure
that the bead did not disconnect
as it was dragged across a surface.
The researchers demonstrated the
effectiveness of their system by suc-
cessfully imaging trial samples with
features as small as 75 nanometers.
While other techniques, such
as near-field scanning microscopy,
can perform sub-diffraction-limit
imaging, they require very expen-
sive systems. The real advantages
of our technique are its simplicity
and its price, says Krivitsky. The
idea could be applied to a variety
of superresolution applications
such as sample inspection, micro-
fabrication and bioimaging.
Imaging:
Microparticles get the
whole picture
Microspheres and a tiny pipette are all it takes to image surfaces with a
resolution below 100 nanometers
A microsphere connected to the end of a pipette enables sub-diffraction-limit imaging.
R
e
p
r
o
d
u
c
e
d

f
r
o
m

R
e
f
.
1

a
n
d

l
i
c
e
n
s
e
d

u
n
d
e
r

C
C

B
Y
-
N
C
-
S
A

3
.0

(
c
r
e
a
t
i
v
e
c
o
m
m
o
n
s
.o
r
g
/
l
i
c
e
n
s
e
s
/
b
y
-
n
c
-
s
a
/
3
.0
/
l
e
g
a
l
c
o
d
e
)

2
0
1
3

L
.
A
.
K
r
i
v
i
t
s
k
y

e
t

a
l
.
1. Krivitsky, L. A., Wang, J. J., Wang, Z.
& Lukyanchuk, B. Locomotion of
microspheres for super-resolution
imaging. Scientific Reports 3,
3501 (2013).
The idea could be
applied to a variety
of superresolution
applications such as
sample inspection,
microfabrication and
bioimaging.
ENGINEERING &
NANOTECHNOLOGY
Laptops have the advantages of
being more versatile and portable
than their desktop counterparts.
But these attributes impose
considerable demands on the
electronic components in a
laptop particularly the hard
drive. The magnetic disk inside
a hard drive rotates at a rate of
several thousand revolutions
a minute. At the same time, a
read/write head moves only a few
nanometers above the disk surface
to access information on the disk.
At such high speeds, large vibra-
tions can permanently damage the
hard drive.
To help reduce hard drive
failures, Jianqiang Mou and
colleagues from the A*STAR Data
Storage Institute in Singapore
have now developed a computer
model that can predict and
minimize the effects of vibrations
on the hard drive and ultimately
help to improve laptop design
1
.
Current designs of many
laptops actually compound the
problems caused by vibrations.
For instance, to provide protec-
tion from external impact and
accidents, laptops are often
encased in special housings
intended to absorb accidental
drops and other shocks. Such
laptop designs can actually be
counterproductive if not done
properly, explains Mou. The
commercial notebook computer
industry rarely understands how
chassis design can substantially
affect the performance of the
hard drive. Some notebook
computers are designed with
vibration sources, for example the
loud speaker, located close to the
hard drive.
To get back to the fun-
damentals of laptop design,
the researchers developed a
theoretical framework that models
the propagation of vibrations from
various components in a laptop,
such as the speakers, to the hard
drive. Underpinning this frame-
work are mathematical equations
that describe the transmission
of vibrations in laptops, and
these equations form the input
for a computer model applied to
specific laptop designs.
The results of the researchers
calculations can be used to
inform general laptop design
strategies. For example, often
very stiff materials are used for
laptop cases to provide enhanced
mechanical strength. However,
stiff materials tend to transmit
high-frequency vibrations more
strongly than f lexible materials,
and it is difficult for hard
drives to compensate for these
frequencies. Softer materials
are preferable as they suppress
higher frequency vibrations,
leaving only slower vibrations
which are easier for hard drives
to compensate.
Our study provides an
effective approach for computer
and hard drive makers to
optimize the chassis design
and component mounting,
adds Mou. Furthermore, the
methodology presented in our
paper can be applied for analysis
and optimal design of other
computer chassis, such as servers
in data centers.
Electronics:
Protecting laptops on the move
A theoretical model for vibrations in laptops provides design strategies for
reducing hard drive failures
Laptop housings are designed to be hard and inflexible to protect laptops from everyday
wear and tear, but a softer, more flexible material may provide a more suitable housing
for suppressing vibrations.

I
n
g
r
a
m

P
u
b
l
i
s
h
i
n
g
/
T
h
i
n
k
s
t
o
c
k
1. Lai, F., Mou, J. Q., See, I. B. L. &
Lin, W. Z. Modeling and analysis
of notebook computer chassis
structure for optimization of
component mounting. International
Journal of Mechanical Sciences 76,
6069 (2013).
Our study provides
an efective approach
for computer and hard
drive makers to optimize
the chassis design and
component mounting.
ENGINEERING &
NANOTECHNOLOGY
Two-dimensional sheets of elec-
tronic materials, such as graphene,
show promise for practical
nanoelectronics applications,
including transparent electronic
circuits used in electronic displays.
Molybdenum disulfide (MoS
2
)
is of particular interest because,
unlike metallic graphene, it is
semiconducting, like silicon the
semiconductor that underpins
todays computer technology.
Now, Yongqing Cai from the
A*STAR Institute of High Perfor-
mance Computing in Singapore,
with colleagues from China and
the United States, has calculated
that, by adding hydrogen to a
MoS
2
surface, regions of the
surface can be converted into
metallic roads. These roads
can transport electrical charges
between different areas of a MoS
2
nanosheet, enabling the fabrication
of integrated electronic circuits
1
.
Computer chips require both
semiconductors and metals. Semi-
conductors (typically silicon) are
the basis for electronic components
such as transistors, whereas metals
(generally copper or gold) are used
for wires that transport electrical
charges around a chip. One
advantage of using two-dimen-
sional sheets such as MoS
2
is that
semiconductors and metals can
be integrated on the same sheet,
facilitating the development
of nanoscale computer chips.
For this to become a reality,
the semiconducting properties of
a MoS
2
sheet need to be modified
to enable some areas of the sheet
to become metallic and hence
electrically conducting. Cai dubs
these regions nanoroads. The
design of conductive nanoroads on
two-dimensional nanosheets in
a way that doesnt compromise
their structural integrity is
critical for transporting electrical
charges and to create reliable,
highly conducting channels for
nanoelectronics applications,
explains Cai.
MoS
2
has to be modified before
it can conduct electricity, since it
requires additional atoms to be able
to transport electrical charges. The
researchers simulated the effects
of adding hydrogen atoms to the
surface of a MoS
2
sheet and found
that MoS
2
will become metallic in
areas where hydrogen atoms bond
to its surface. They showed that
adding lines or chains of hydrogen
atoms to the surface created
metallic strips. The researchers
calculations reveal that these strips,
or nanoroads, are reliable electrical
conductors, and, importantly, they
do not damage the structure of the
underlying sheets.
In terms of practical imple-
mentation, the technology already
exists for depositing hydrogen
on semiconductor nanosheets:
hydrogen has been deposited on
other two-dimensional sheets,
including graphene. Before MoS
2

sheets can be used to produce
components such as transistors, a
method for producing electron-
deficient regions needs to be
developed. Once this practical
challenge has been addressed,
the way will be open to success-
fully using MoS
2
in integrated
electronic applications.
Nanotechnology:
Paving the way for
electronic applications
The formation of electrically conducting nanoroads on atomically thin
semiconductor nanosheets enables the integration of electronic components
Conducting nanoroads on the surface of nanosheets of molybdenum disulfide could
underpin integrated electronics on this ultrathin material.

c
h
u
y
u
/
i
S
t
o
c
k
/
T
h
i
n
k
s
t
o
c
k
1. Cai, Y., Bai, Z., Pan, H., Feng, Y. P.,
Yakobson, B. I. & Zhang, Y.-W.
Constructing metallic nanoroads on
a MoS
2
monolayer via hydrogenation.
Nanoscale 6, 16911697 (2014).
ENGINEERING &
NANOTECHNOLOGY
As individual computer users
increasingly access the Internet
from different smartphones,
tablets and laptops, many are
choosing to use online cloud
services to store and synchronize
their digital content. Cloud
storage allows consumers to
retrieve their data from any
location using any device and
can provide critical backups in
the case of hard disk failure. But
while people are usually vigilant
about enacting security measures
on personal computers, they often
neglect to consider how safe their
files are in the cloud.
Now, findings from a team
led by Jianying Zhou of the
A*STAR Institute for Infocomm
Research in Singapore promise to
improve the security of popular
online services and better protect
users by revealing hidden f laws
associated with an important
cloud storage feature the
ability to share files with friends,
co-workers or the public
1
.
Sharing content is an attractive
way to let far-f lung colleagues
view and collaborate on projects
without using email attachments,
which often have strict file size
limitations. Data sharing can be
either public, with no access con-
trols; private, in which the cloud
service provider authenticates
sharing through login controls; or
secret uniform resource locator
(URL) sharing where people
without an account on the cloud
service can access data by fol-
lowing a specific web link.
The A*STAR-led researchers
analyzed the security of three
well-known cloud service pro-
viders Dropbox, Google Drive
and Microsoft SkyDrive
and found that all three had
vulnerabilities many users might
encounter. They uncovered several
risks related to the sharing of
secret URLs. Because URLs are
saved in various network-based
servers, browser histories and
Internet bookmarks, frequent
opportunities exist for third
parties to access private data.
Furthermore, the URL recipient
may send the link to others
without the data owners consent.
Another danger lies in the
practice of URL shortening
reducing long web addresses to
brief alphanumeric sequences for
easier sharing on mobile devices.
Although the original URL may
point to a privately shared file,
shortening changes this address
into plain text unprotected by
encryption. Zhou also notes that
because short URLs have very
limited lengths, they are suscep-
tible to brute-force attacks that
can dig out supposedly secret files.
Zhou explains that the root
cause of cloud security problems
lies in the need to balance
usability with privacy protection.
Users should be careful when
they share files in the cloud
because no system is perfectly
secure. The cloud industry,
meanwhile, needs to constantly
raise the bar against new attacks
while keeping the service as
functional as possible.
Cybersecurity:
Cracks emerge in the cloud
A systematic analysis reveals that cloud storage services have security
weaknesses that can inadvertently leak users data
Storing digital content on cloud computing services is becoming increasingly popular but presents security concerns when users try
to share their data with others.

D
i
v
V
e
c
t
o
r
/
i
S
t
o
c
k
/
T
h
i
n
k
s
t
o
c
k
1. Chu, C.-K., Zhu, W.-T., Han, J.,
Liu, J. K., Xu, J. & Zhou, J. Security
concerns in popular cloud storage
services. IEEE Pervasive Computing
12, 5057 (2013).
ENGINEERING &
NANOTECHNOLOGY
In 2009, the global number of
city dwellers surpassed that of
rural dwellers. Understanding
how cities evolve is vital to a
world that will continue to
urbanize. Now, researchers
at A*STAR have developed
a computer model that can
reconstruct cities building
them from the bottom up to
investigate the fundamental
mechanisms that underpin and
govern city growth
1
.
Modeling dynamic urban
growth presents many challenges
due to the complexity of city
systems and the technical and
data requirements of model
building. Christopher Monterola
and co-workers at the A*STAR
Computing, together with scien-
tists in the United Kingdom, have
built a model based on a so-called
cellular automation system, which
uses a minimalist approach to
simulate city growth and support
planning for sustainable cities.
The overarching vision
of our team is to capture the
form, structure and dynamics
of different cities to better
understand, manage, design and
evaluate urban systems, explains
Monterola. Essentially, we are
trying to generate the recipe for a
sustainable, smart city.
The model works by taking
a simple set of rules for
example, land value and physical
constraints to building such as
water bodies and parklands
and defining the probable land
use of each cell, or unit of land,
according to the information
provided by its neighboring
cells. The model then takes into
account the different land-use
sectors industrial, business
and residential and determines
their range of inf luence, to
decide how far a certain type
of land use will spread within a
certain radius. For each simula-
tion, the team set the model
center at the original marketplace,
or central business district, of a
real city, and let the model grow
the city from there.
The team validated their
results using high-resolution data
from various cities, including Sin-
gapore, Toronto and Las Vegas.
Their model replicated, fairly
accurately, the land-use patterns
of the actual cities (see image).
Our results suggest that there
are some generic rules that a
growing city follows as it evolves,
Monterola states. We found
that there was an effective and
stable cluster size for business,
residential and industrial areas
in all the cities studied, and their
size ratios are remarkably regular.
Hence, sustainability concepts
must be somehow anchored on
accepting this innate evolution,
and policies need to be planned
around such constraints.
The team plans to further
develop their urban growth
model, for example by
investigating the limitations
on individual city transport
systems and working on ways to
make cities run more efficiently
and sustainably.
Urban planning:
Modeling city growth from the
ground up
A computer model that can replicate the growth of cities has valuable
implications for urban planning and sustainability
Singapore
Toronto
Simulations generated by a model that replicates city growth to determine likely
land-use patterns (right) show close alignment between actual land-use maps (left) for
two sample cities Singapore (top) and Toronto (bottom).
M
o
d
i
e
d

f
r
o
m

R
e
f
.
1

a
n
d

l
i
c
e
n
s
e
d

u
n
d
e
r

C
C

B
Y

4
.0

(
c
r
e
a
t
i
v
e
c
o
m
m
o
n
s
.o
r
g
/
l
i
c
e
n
s
e
s
/
b
y
/
4
.0
/
l
e
g
a
l
c
o
d
e
)

2
0
1
3

J
.
D
e
c
r
a
e
n
e

e
t

a
l
.
1. Decraene, J., Monterola, C.,
Lee, G. K. K., Hung, T. G. G., &
Batty, M. The emergence of
urban land use patterns driven
by dispersion and aggregation
mechanisms. PLoS ONE 8,
e80309 (2013).
The overarching vision
of our team is to capture
the form, structure
and dynamics of
diferent cities to better
understand, manage,
design and evaluate
urban systems.
The Agency for Science, Technology and Research (A*STAR) is Singapores lead government agency dedicated to
fostering world-class scientic research and talent for a vibrant knowledge-based economy.
A*STAR actively nurtures public-sector research and development in biomedical sciences, physical sciences
and engineering, and spurs growth in Singapores key economic clusters by providing human, intellectual and
industrial capital to our partners in industry and the healthcare sector.
A*STAR currently oversees the following research institutes, consortia and centers and supports extramural
research with universities, hospital research centers, and other local and international partners.
ASTAR_Collection_2012_09_Dividers.indd 8 28/09/2012 12:25PM
Bioinformatics Institute (BII)
Bioprocessing Technology Institute (BTI)
Clinical Imaging Research Centre (CIRC)
Data Storage Institute (DSI)
Experimental Therapeutics Centre (ETC)
Genome Institute of Singapore (GIS)
Institute of Bioengineering and Nanotechnology (IBN)
Institute of Chemical and Engineering Sciences (ICES)
Institute of High Performance Computing (IHPC)
Institute for Infocomm Research (I
2
R)
Institute of Materials Research and Engineering (IMRE)
Institute of Medical Biology (IMB)
Institute of Microelectronics (IME)
Institute of Molecular and Cell Biology (IMCB)
National Metrology Centre (NMC)
Singapore Bioimaging Consortium (SBIC)
Singapore Institute for Clinical Sciences (SICS)
Singapore Institute of Manufacturing Technology (SIMTech)
Singapore Immunology Network (SIgN)
www.astar-research.com

A STAR Research April-September 2014

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

A STAR Research April-September 2014

Uploaded by

Copyright:

Available Formats

HIGHLIGHTING THE BEST OF A STAR

You might also like